Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis (2023)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

B-1   Literature Review Contents B-3 1 Introduction B-3 1.1 Identifying Documents for Review B-4 1.2 Key Findings in the Literature B-4 2 Risk and Resilience Policies in Transportation B-6 2.1 Federal Transportation Risk and Resilience Initiatives B-10 2.2 International Requirements for Risk-Based Asset Management B-11 3 Risk and Resilience Definitions B-11 3.1 Risk Definitions B-11 3.2 Resilience Definitions B-13 3.3 Relationship Between Risk and Resilience B-15 3.4 Frameworks for Risk and Resilience B-18 4 Methods of Risk and Resilience Assessment, Metrics, and Performance Indicators B-18 4.1 Risk Assessment Methodologies and Metrics B-18 4.1.1 Qualitative Risk Assessment Methods B-21 4.1.2 Quantitative Risk Assessment Methods B-24 4.1.3 Hybrid Risk Assessments Methods B-26 4.1.4 Risk Metrics B-28 4.1.5 Risk-Based Performance Measures B-30 4.2 Resilience Assessment Methodologies and Metrics B-35 5 Major Components of a Risk Assessment Methodology B-35 5.1 Methods to Identify Critical Transportation Assets B-36 5.2 Threat Assessment and Incorporation of Uncertainty B-37 5.2.1 Threat Identification B-37 5.2.2 Data Sources to Support Threat Modeling B-38 5.2.3 Types of Threat Modeling B-48 5.3 Estimating Vulnerability of Assets from Relevant Threats or Hazards B-52 5.4 Types of Consequences from Identified Threats B-55 6 Tools for Risk and Resilience Assessment B-55 6.1 Spreadsheet Models B-56 6.2 Software Tools B-58 6.3 Web-Based Application Tools B-60 6.4 Methods for Developing Recovery Strategies A P P E N D I X B

B-2 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis B-61 6.5 Methods for Risk Management B-62 6.6 Methods to Improve System Resilience B-63 7 Tracking Mechanisms for Risk and Resilience Metrics B-63 7.1 Recurring Reports B-64 7.2 Story Maps B-64 7.3 Dashboards B-64 8 Barriers and Constraints for Implementation of Quantitative Risk and Resilience Assessment Methods

Literature Review B-3 1 Introduction This project’s primary objectives are to (1) develop a comprehensive and consistent set of risk and resilience terminology and (2) develop a risk and resilience framework and formulate a research roadmap to develop a highway manual that supports quantitative all-hazard risk and resilience assessments for state and local DOTs. These objectives will be met through the conduct of four tasks. Task 1 includes three activities: (a) develop a risk and resilience glossary of terms, (b) prepare a literature review, and (c) conduct a gap assessment. This appendix is the second of the three Task 1 activities. This appendix summarizes established practices, innovative approaches, and research efforts related to the risk and resilience analysis of highway assets. Section 1.1 contains a summary of the process applied to identify relevant literature, while Section 1.2 provides a summary of findings. Sections 2 through 8 delve into specific aspects of risk and resiliency, including the following: • Section 2: Risk and Resilience Policies in Transportation • Section 3: Risk and Resilience Definitions • Section 4: Methods of Risk and Resilience Assessment, Metrics, and Performance Indicators • Section 5: Major Components of a Risk Assessment Methodology • Section 6: Tools for Risk and Resilience Assessment • Section 7: Tracking Mechanisms for Risk and Resilience Metrics • Section 8: Barriers and Constraints for Implementation of Quantitative Risk and Resilience Assessment Methods The appendix is based on a thorough, comprehensive literature review related to risk and resilience analysis approaches of highway assets, including policies in transportation, perfor- mance indicators, guidance and frameworks, and tools. 1.1 Identifying Documents for Review To identify documents for review, the research team first compiled a working list of 230 active and past research reports where risk and resilience were discussed or utilized in highway asset analysis. This list is in Appendix A and includes general research reports, state asset manage- ment plans, guidelines, state performance reports, management tool application documents, and state policies and recommendations. While transportation is the central focus of this project, literature from other relevant fields was included for added insight. The search included the following: • Traditional academic journal databases via Google Scholar • TRID: TRB’s integrated database • AASHTO’s TAM Portal • Review of U.S. DOT website as well as state DOT websites and publications • Materials suggested by resiliency and risk management experts. The research team reviewed the list of 230 documents, with summary comments on the docu- ments’ relevance included in a table of resources. The resource review was organized by section in a shared Excel document so that the research team could provide input. This allowed easy access and comprehension of each topic covered in the literature review. Each resource had a link to access the complete publication, the type and date of publication, comments relating to the document, and the name of the contributing research team member. The format of this review allowed the team to pull the necessary resources easily and efficiently when producing the literature review. The table was available to the NCHRP Project 23-09 panel through a shared drive. From this list, 219 documents were deemed relevant and helpful. These form the basis of the literature review.

B-4 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis 1.2 Key Findings in the Literature The research team made the following key observations about the findings in the literature review: • In many cases, risk and resilience concepts are used interchangeably among state and local DOTs. There is a need for education to help transportation professionals understand how these two terms relate and the relevant metrics for each term. • It has been observed that many metrics for resilience incorporate the use of risk measures. There is an inverse relationship between risk and resilience where reducing risk increases resilience and vice versa. However, other factors such as planning for a response, recovery, and adaptation play significant roles in system resilience. • There are minimal differences in AASHTO and FHWA definitions for risk and resilience. Many states have adopted either the FHWA or the AASHTO risk and resilience definitions. However, some states have developed their own definitions. • Most states are currently performing risk assessments by using simple risk registers based on a five-point rating that relies on a collective judgment. A few state DOTs are expanding their initiatives and incorporating quantitative methodologies for risk and resilience assessments. • Multiple federal transportation agencies, state DOTs, and metropolitan planning organi- zations (MPOs) have performed vulnerability assessments using the FHWA Vulnerability Assessment and Adaptation Framework (VAAF) or a variation of it during FHWA-sponsored resilience pilot projects. • There is no standard methodology for performing quantitative risk and resilience assessments for transportation agencies; however, other sectors such as the wastewater industry and the energy industry have developed methods that are more widely adopted and applied in their industries. 2 Risk and Resilience Policies in Transportation Risk and resilience policies within transportation agencies have been mostly informal for decades but are currently being formalized and expanded on in many areas, including being integrated into the core agency operating strategy. The process began with the implementation of the Moving Ahead for Progress in the 21st Century Act (MAP-21) (2012), enacted in 2012, followed by the Fixing America’s Surface Transportation Act (FAST Act) (2015), enacted in 2015. These laws effectively required DOTs to enact risk management policies, localizing them in each DOT’s Transportation Asset Management Plan (TAMP). MAP-21 introduced the concept of a TAMP, enforcing the importance of integrated and optimized asset, performance, and risk management. Reforms to the transportation planning process included “incorporating performance goals, measures, and targets into the process of identifying needed transportation improvements and project selection [The Moving Ahead for Progress in the 21st Century Act (MAP-21)].” The FAST Act expanded on MAP-21, providing long-term funding for surface transportation infrastructure planning and investment. FHWA regulations following the FAST Act introduced performance measures and thresholds, with the performance measures based on three categories, as shown in Table B-1. It was expected that DOTs would perform risk management through their TAMPs by iden- tifying risks, prioritizing them, and presenting mitigation strategies. Many states created “risk registers” to accomplish the first two steps. For example, the California Department of Trans- portation (Caltrans) created a risk management handbook that outlines guidelines and consid- erations for a risk register. It recommends a risk register for any Caltrans project, but it requires one if the project cost is over $1 million. The handbook includes three levels of risk register

Literature Review B-5 details that may apply to a project contingent on several variables, primarily project cost (Caltrans 2018). Additional factors to consider when determining what level of risk analysis and manage- ment should be included are as follows: • Political sensitivity • Type of project • Location of the project and the community it serves • Duration of the project • Project stakeholders • Project sponsor’s sensitivity to changes in project schedule/cost • Level of scoping and preliminary planning that has been done previously In 2017, CFR Title 23 Part 515, deemed the asset management rule, was put in place, stating that state DOTs shall “develop a risk-based asset management plan that describes how the National Highway System (NHS) will be managed. . . .” This included establishing a process for conducting performance gap analysis, life-cycle planning, development of a risk management plan, development of a financial plan, and development of investment strategies at minimum. According to the Code of Regulations Title 23, Part 450.306(b)(9), the metropolitan trans- portation planning process should address the improvement of “resiliency and reliability of the transportation system . . .” (Code of Federal Regulation: https://www.ecfr.gov/cgi-bin/text- idx ?c=ecfr;sid=e2662fc63c225d496d1fa6ce22ea6cb8;rgn=div5;view=text;node=23%3A1.0.1.5.11; idno=23;cc=ecfr#se23.1.450_1306). As a result of this policy, DOTs should consider resilience as a planning factor when assessing projects, strategies, and services. In addition to the federal requirement of developing risk-based asset management plans, MAP-21 requires state DOTs to conduct periodic evaluations of roads and bridges that have needed repairs or reconstruction on two or more occasions due to catastrophic events and decide whether there are suitable alternatives. This requirement was established to conserve federal resources and promote public safety. Specific deadlines for implementing this policy were published in 23 CFR Section 667. States had to do the first review by November 23, 2018, and then update the reviews every 4 years as needed. Follow-on reviews had to be completed by November 23, 2020. These reviews must be summarized in TAMPS and integrated into trans- portation plans and programs (Allen 2020). As per 23 CFR Section 551, the periodic reviews described in 23 CFR Section 667 require TAMPs to incorporate the following guidance. Specifi- cally, by April 2018, TAMPS should: • Establish a planning process for the entire life cycle of assets that considers current and future conditions, i.e., climate change, extreme weather events, seismic events, etc. [23 CFR 515.7(b)]. Measures Metrics for Performance Measures PM1 Safety Number and rate of fatalities per 100 million vehicle miles traveled (VMT) Number and rate of serious injuries per 100 million VMT Number of non-motorized fatalities and serious injuries PM2 Infrastructure Condition Percentage of pavement in good and poor condition (Interstate and NHS) Percentage of bridge deck area in good and poor condition (Interstate and NHS) PM3 System Reliability Percentage of reliable person miles traveled (Interstate and NHS) Truck Travel Time Reliability Index (TTTR) Total emissions reduction for applicable criteria pollutants Table B-1. FHWA required performance measures.

B-6 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis • Establish a risk-based asset management plan that includes risk assessments that address current and future conditions, address reoccurring damage and the associated costs, estimate the likelihood of risks, prioritize risks, and develop a mitigation plan for the highest-priority risks [23 CFR 515.7(b)]. Between the 51 states’ TAMPs, there are many methods to implement risk and resilience strategies. The formal discussion around emerging standard policies being utilized has established that DOTs begin by setting the standard for what is expected. From there, informal norms have been developed and serve as the main source of guidance to improve and enhance asset management practices. Many states have built on the policies, tailoring definitions and requirements to serve their needs best. Different efforts have been made to help agencies implement risk and resilience analysis in transportation activities. 2.1 Federal Transportation Risk and Resilience Initiatives Federal, state, and local governments invest in their transportation infrastructure daily. Trans- portation is a long-term investment with many inherent risks to that daily investment including natural disasters, operational and funding gaps, asset condition and performance and more. The industry also lacks standard methods in assessing risk and determining the appropriate distinction between the commonly recognized “negative” risks instead of the justified and even necessary risks involved in optimizing the systematic efficiency. A current example of this dilemma is considering connected and/or autonomous vehicle tech- nologies on the federal highway system. The current state-of-the-industry indicates enhanced traffic control infrastructure would improve the safety and operational performance of these technologies; however, this added investment only benefits technologies that are largely yet unproven. However, the risk in not accommodating those emerging technologies can have sig- nificant impacts on the local economy, operations, and asset condition. Considering the poten- tial impacts has led FHWA to include the State Transportation Innovation Council concept for each state to implement as part of its Every Day Counts Initiative (FHWA 2020b). Natural risks also encourage strategic initiatives for risk mitigation. From record setting hurricanes along the south and east coasts, wildfires ravaging throughout California, flooding in Colorado, and ice storms in Atlanta, assets are being disrupted everywhere (FHWA n.d.a). Transportation agencies need to ensure their infrastructure is resilient and prepared for any potential risks to address these issues. Being more resilient is at the top of the list for many ini- tiatives in the coming years. This critical issue was included in the U.S. DOT Strategic Plan for Fiscal Year 2018–2022, where the development of new tools to improve durability and resilience will be a priority. Doing this will extend the life span of essential infrastructure across the United States, but it will also reduce future costs. In addition to strategic planning, the FHWA put together a white paper project providing a baseline of efforts DOTs and MPOs have already implemented to integrate resilience, as well as a handbook to guide practitioners at all resilience planning levels on how to better integrate resil- ience into their agencies (FHWA n.d.c). Many states have cited their incorporation of resilience measures as due to federal regulations, as seen in Table B-2 or local state policies. Another effort by the federal government to assist in the incorporation of resilience mea- sures is the ability for states to use Emergency Relief (ER) Program funds on improving long- term resilience on federal-aid highways, as long as it is consistent with current standards and

Literature Review B-7 Effective Date Overview Source June 27, 2016 “(a) Each State shall carry out a continuing, cooperative, and comprehensive statewide transportation planning process that provides for consideration and implementation of projects, strategies, and services that will address the following factors: … (9) improve the resiliency and reliability of the transportation system and reduce or mitigate stormwater impacts of surface transportation.” 23 CFR 450.206(a) June 27, 2016 “(b) The metropolitan transportation planning process shall be continuous, cooperative, and comprehensive, and provide for consideration and implementation of projects, strategies, and services that will address the following factors: … (9) Improve the resiliency and reliability of the transportation system and reduce or mitigate stormwater impacts of surface transportation;” 23 CFR 450.306(b) Long-range statewide transportation plan adopted after May 2018 meets requirements “(c) The long-range statewide transportation plan shall reference, summarize, or contain any applicable short-range planning studies; strategic planning and/or policy studies; transportation needs studies; management systems report; emergency relief and disaster preparedness plans;” 23 CFR 216(c) On or after May 27, 2018, an MPO meets requirements to adopt a metropolitan transportation plan “(f) The metropolitan transportation plan shall, at a minimum, include: … (7) Assessment of capital investment and other strategies to preserve the existing and projected future metropolitan transportation infrastructure, provide for multimodal capacity increases based on regional priorities and needs, and reduce the vulnerability 23 CFR 450.324(f)(7) of the existing transportation infrastructure to natural disasters.” October 2, 2017 Asset Management Plan (c) A State DOT shall establish a process for developing a risk management plan. This process shall, at a minimum, produce the following information: … (6) Risk management analysis, including the results for NHS pavements and bridges, of the periodic evaluations under part 667 of this title of facilities repeated damaged by emergency event. … and (h) A State DOT shall integrate its asset management plan into its transportation planning processes that lead to the STIP, to support its efforts to achieve the goals in paragraphs (f)(1) through (4) of this section. 23 CFR 515.7 (c)(6) and 515.9 (h) Mandatory and due by November 23, 2018 State DOTs must evaluate facilities that have repeatedly been damaged in emergency events. FAST Act 23 CFR 667 Nonbinding The National Highway Freight Program has a goal to “improve the … resiliency of freight transportation in rural and urban areas.” FAST Act Nonbinding Goals for the national transportation system include increasing safety, security, and reliability. MAP-21 Nonbinding National Infrastructure Protection Plan invests to produce significant reductions in national risk. Department of Homeland Security Table B-2. Federal laws and regulations that require resilience considerations.

B-8 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis economically justied (FHWA, Kalla, and Shepherd n.d.). is allows states to bounce back from disasters and create stronger, more resilient assets for any future impacts. We are still learning the full extent of how the impacts of natural disasters and climate change that are constantly threatening critical infrastructure aect transportation systems. ere are many proactive measures being taken by states, but there is still an opportunity to learn more and integrate even more resilient eorts. A study shows four pathways of disruption to transportation systems as a result of climate change and extreme weather events, as seen in Figure B-1 (Markolf et al. n.d.). Infrastructure systems are only becoming more complex and interconnected, along with the constantly evolving challenges and threats. e movement toward gaining a better understanding and implementing eective measures is crucial to creating a more resilient infrastructure. e U.S. Government Accountability Oce took a closer look at the impact of climate change on coastal communities and areas impacted by environmental risks (USGAO 2020). e plea was made in Congress to consider establishing a federal pilot program solely dedicated to cli- mate migration. Climate migration is the action of retreat or relocation from vulnerable areas to combat the eects of climate change on many communities. Current federal programs provide limited support, and assistance has been very slow for areas in need. Implementing a program focused on supporting impacted communities will ultimately enhance the nation’s climate resil- ience and reduce federal scal exposure. Risk and resilience eorts in U.S. transportation agencies have been occurring informally for decades. A formalized practice began with MAP-21 and the FAST Act as well as many other regulations that can be seen in Table B-2. ese initial federal authorization bills guided the development of state TAMPs, a requirement for all DOTs to ensure risk management is being considered. e TAMPs showcase the various informal norms and standards developed and used over the years while implementing risk and resilience measures. While standard denitions have evolved from FHWA and AASHTO, many states have certain objectives and practices specic to their states’ needs and tailor their understanding of risk and resilience to better align with their initiatives. e FHWA has a Final Rule for risk management processes, as seen in Figure B-2 that sets the general guidance and approach for the state (Caltrans 2019). In 2017, the FHWA developed Incorporating Risk Management into Transportation Asset Management Plans as interim guidance. e document “guides the risk element of the (Adapted from Markolf et al. n.d.) Figure B-1. Four pathways of disruption to the transportation system.

Literature Review B-9 TAMP, defines risk, and provides guidance on how the risk element can be applied to meet risk-based TAMP requirements” (FHWA 2017). The guidance assists agencies in devel- oping their TAMPs, encouraging the use of risk and resilience practices throughout all departments. Multiple NCHRP reports have dissected the deployment of risk and resilience practices in state DOTs, developing guidance and tools for their incorporation. NCHRP Project 20-117 created a Transportation Resilience Guide and Toolkit and held a national summit and peer exchange on transportation resiliency (Hammond n.d.). The creation and implementation of these tools helps states deploy greater risk and resilience initiatives, learning from other trans- portation agencies and practitioners. The final report of NCHRP Project 08-93, “Managing Risk across the Enterprise: A Guide for State Departments of Transportation,” was made available in 2016 and provided a “comprehensive framework to identify and manage risk” for state DOTs (Proctor et al. 2016). The guidance lays out a step-by-step process for practitioners and agencies (Source: Caltrans 2019.) Figure B-2. Risk management process defined by FHWA Asset Management Final Rule.

B-10 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis to add risk management and resilience measures into already incorporated management pro- cesses. This concept can be seen in Figure B-3 (Proctor et al. 2016). Another report, “Incorporating Resilience into Transportation Planning and Assessment,” explored incorporating resilience into long-term transportation planning efforts by state DOTs and MPOs (Weilant et al. 2019). A framework was developed using interviews with stakeholders at DOTs and planners in MPOs and reviews of published literature to create a logic model for incorporating resilience methods. It was developed in the interest of DOT and MPO planners, with the intent to be used to “modify long-term transportation planning processes to better incorporate resilience.” While there is no specific national guidance, there are many resources available to guide transportation agencies in various directions, depending on their wants and needs. The devel- opment of new methods and practices is constantly evolving, and the implementation of risk management and resilience measures is more important now than ever. 2.2 International Requirements for Risk-Based Asset Management Maximizing return while also being environmentally responsible and safe is more critical than ever before. Aligning these can be tricky, and misalignment may cost time, funds, and resource waste. To combat this, the International Organization for Standardization (ISO) 55000 was cre- ated in 2014. This standard took the place of PAS 55 Asset Management, a similar standard with a key difference that PAS 55 focused on physical assets only. ISO 55000 is a family of standards providing three sets of information to assist risk-based asset management (Sanford 2015). • ISO 55000: Asset Management—Overview, principles, and terminology • ISO 55001: Asset Management—Management systems—Resources • ISO 55002: Asset Management—Management systems—Guidelines for the application of ISO 55001 These three standards are essential because they “represent a global consensus on what asset management is and what it can do to increase the value generated by all organizations” (Sanford 2015). The standards clarify organizational context, define leadership and support roles, specify performance evaluation elements and other key components to successfully manage risk. (Source: Proctor et al. 2016.) Figure B-3. Illustration that risk management and performance management operate as parallel.

Literature Review B-11 3 Risk and Resilience Definitions Risk and resilience terms have been used interchangeably and applied in different sectors and fields of study; however, they have different definitions and measures. This section presents an overview of some of the definitions and metrics used in the transportation sector for each term. 3.1 Risk Definitions Many definitions of risk exist in many sectors. Most colloquial definitions of risk share the same elements, including uncertainty or probability, an unpleasant event, and injury or damage. The ISO’s 31000 guide to risk management states that “Risk is often expressed in terms of a combination of the consequences of an event (including changes in circumstances) and the associated likelihood of occurrence” (International Organization for Standardization 2018). The ISO’s guide to risk assessment for structures adds the additional dimension of severity: “Risk is a combination of the probability or frequency of occurrence of an event and the magnitude of its consequence” (International Organization for Standardization 2018). Another definition of risk is presented by the American Society of Mechanical Engineers (ASME) guidebook to the Risk Analysis and Management for Critical Asset Protection (RAMCAP Plus) method for risk assessment. RAMCAP defines risk as “the potential for loss or harm due to an untoward event and its adverse consequences” (ASME 2009). In the transportation sector, AASHTO states in its Guide for Enterprise Risk Management that risk “involves more than just threats or hazards” and defines risk as “the positive or negative effects of uncertainty or variability on agency objectives” (AASHTO 2016a). The FHWA asset management rulemaking (23 CFR 515.5) provides the same definition for risk as AASHTO’s Guide for Enterprise Risk Management. Transportation agencies frequently adopt this definition, and it has been included in multiple state DOTs TAMPS, including CDOT (2019), Kansas DOT (KDOT 2019), and Minnesota DOT (MnDOT 2019). In addition, many transportation agencies have adopted FHWA’s risk definition stated on 23 CFR 515.5 or devel- oped their own definition. Table B-3 presents some standard risk definitions. Many definitions of risk share notions of probability and adverse effects and how risk is calculated by multiplying uncertainty times a measure of consequences. 3.2 Resilience Definitions Similar to the definition for risk, there is also no universal definition of resilience or resiliency. Many definitions have evolved in different sectors and fields of study (see Table B-4). However, there are many parallel definitions found within the literature. Some highly cited definitions for resilience outside the transportation sector include a defi- nition from the U.S. Department of Homeland Security (DHS) which defines resilience in the 2009 edition of the National Infrastructure Protection Plan (NIPP) (DHS 2009) as “The ability to resist, absorb, recover from, or successfully adapt to adversity or a change in conditions.” Weilant et al. (2019) expand on these themes: • “Reducing the likelihood of a disaster and increasing the ability of a community to absorb or resist a shock.” • “Increasing the adaptability of a system while maintaining functions in the presence of a shock.” • “Reducing the time to recovery to normal functioning, which might be different from pre-event functioning.”

B-12 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis Similarly, the National Academy of Sciences (NAS) defines resilience as “the ability to plan and prepare for, absorb, recover from, and adapt to an adverse event.” (The National Academies 2012). A compiled collection of similar definitions from researchers within the transportation sector include (Freckleton et al. 2012): • “A system’s ability to maintain its demonstrated level of service or restore itself to that level of service in a specified time frame” (Serulle et al. 2011). • “A characteristic that enables the system to compensate for losses and allows the system to function even when infrastructure is damaged or destroyed” (Battelle 2007). • “A system’s ability to accommodate variable and unexpected conditions without catastrophic failure” (Litman 2007). • “A system’s ability to absorb the consequences of disruptions to reduce the impact of disrup- tions and maintain freight mobility” (Ta et al. 2009). Definitions of resilience from outside the transportation sector emphasize the ability to withstand or recover from a disruption. In 2009, the AASHTO-TRB Transportation and Security Summit proposed the following defini- tion: “The ability of a system to provide and maintain an acceptable level of service or functionality in the face of major shocks or disruptions to normal operations” (AASHTO 2016b, Flannery et al. 2018). However, the most recent definition from AASHTO’s Special Committee on Transporta- tion Security and Emergency Management (SCOTSEM) states that resilience is “the ability to prepare for, absorb, recover from, or more successfully adapt to adverse events” (AASHTO 2016b). Further, FHWA Order 5520 defines resilience as “the ability to anticipate, prepare for, and adapt to changing conditions and withstand, respond to, and recover rapidly from disruptions.” Many transportation agencies have adopted FHWA’s definition or developed similar defini- tions for their purposes, as illustrated in Table B-4. An FHWA white paper pointed out that most state DOT and MOP resilience definitions focus on “the ability to prepare for and recover from disasters and disruptive events” but vary on “how agencies propose to build that ability with Definition Source An uncertainty that can have either positive or negative impacts Colorado Department of Transportation, Risk and Resilience Analysis Procedure 2020 A combination of the likelihood that an asset will experience a climate impact and the severity or consequence of that impact Filosa et al. 2017 The potential for loss or harm due to an untoward event and its adverse consequences ASME 2009 Effect of uncertainty on objectives ISO 31000, 2009 The positive or negative effects of uncertainty or variability on agency objectives AASHTO 2016b A combination of the magnitude of potential consequence(s) of climate change impacts(s) and the likelihood that the consequence(s) will occur. National Research Council 2010 “Measure of the probability and severity of an adverse effect to life, health, property, or the environment.” "Probability of an adverse event times the consequences if the event occurs." ISSMGE TC32: Technical Committee on Risk Assessment and Management Glossary of Risk Assessment Terms, Version 1, July 2004. Table B-3. Examples of risk definitions.

Literature Review B-13 some emphasizing adaptive capacity and robustness, while other prioritize swiftness in recovery response (Dix et al. 2018).” As part of this project, a glossary of terms was developed with a more extensive list of risk and resilience terminology. 3.3 Relationship Between Risk and Resilience Risk and resilience terms are often used interchangeably due to misconception. To better understand the relationship between risk and resilience, a review of some common definitions is in order. Risk can be described as an event that may cause either harm or improve a situation (opportunity) and is associated with some probability of occurrence. Risk management, on Transportation Agency Resilience Definition Anchorage Metropolitan Area Transportation Solutions “resilience means how to work around outcomes to get back up running quickly. ” Arkansas DOT “resilience also implies transformation, so not only is the infrastructure serviceable to survive or recover but it can adapt to a changing environment in which it operates." Baltimore Regional Transportation Board states that resilience means its system is "better able to adapt to a variety of potentially significant future changes." CDOT “Resilience is the ability to keep our roads open and functional in the face of unexpected events and challenges.” “The ability of a system to rebound, positively adapt to, or thrive amidst changing conditions or challenges, including human-caused and natural disasters, and to maintain quality of life, healthy growth, durable systems, economic vitality, and conservation of resources for present and future generations.” MnDOT “reducing vulnerability and ensuring redundancy and reliability to meet essential travel needs.” Northeast Ohio Areawide Coordinating Agency (NOACA) “Resiliency is a process for managing complex infrastructures rather than a single outcome… As such, a resiliency framework takes an adaptive life-cycle approach to tackling the dynamic challenges that confront today's complex infrastructure systems. Embedded in it is the capability to protect its assets, anticipate and detect threats, prevent risks of known failures, withstand unanticipated disruptions, and respond and recover rapidly when the worst does happen.” Rockingham Planning Commission “a capability to anticipate, prepare for, respond to, and recover from significant multi-hazard threats with minimum damage to social well-being, the economy, and the environment.” San Francisco Metropolitan Transportation Commission (MTC) “includes a desire to ‘enhance climate protection and adaptation efforts’ in its definition of resilience.” Tennessee “The ability of the transportation system to withstand and recover from incidents.” Wisconsin DOT “a resilient transportation system is able to quickly respond to unexpected conditions and return to its usual operational state.” Table B-4. Examples of resilience definitions used by transportation agencies.

B-14 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis the other hand, is an action taken to mitigate risk. Typical risk management strategies include: (1) avoid, (2) reduce, (3) tolerate, (4) transfer, and (5) take advantage of (i.e., “don’t let a good crisis go to waste.”). Several example denitions of resilience include the following: • FHWA: “to anticipate . . . and adapt to changing conditions, and withstand, . . . and recover rapidly from disruptions.” • National Research Council: “. . . absorb, recover from, or more successfully adapt to adverse events.” • From Ecology: “. . . the ability of a system to resist and recover from a disturbance” (Hodgson et al. 2015). Common themes found in denitions of resilience include resisting and recovering from a dis- ruption. ere is an inverse relationship between risk and resilience. Natural and human-made threats can negatively aect a system’s resilience. A threat or hazard can disrupt a system and impede recovery. e risk associated with a threat can be measured using the probability of an adverse event with an unknown severity causing an unknown degree of damage, disruption, or service loss. On the other hand, an agency can take the opportunity to reduce the likelihood of an adverse event or mitigate its impacts, thus increasing system resilience. As Flannery et al. (2018) state, “when risk or the expected nancial loss decreases, resilience increases. By reducing vulner- ability to a threat (moving toward a value of 0), resilience is highest, and the expected risk tends toward zero.” is relationship between risk, resilience, and vulnerability is visualized in a graphic prepared by the DHS (see Figure B-4) (DHS 2010). In the graph, risk and resilience are inversely related. As vulnerability increases, so does risk. Resilience, in this case, is measured in terms of the time it takes a system to return to normal functioning, i.e., performance-time units. As vulnerability increases, resilience decreases, and the time it takes to restore a system to full functioning is longer. Even though risk and resilience have dierent meanings, they are interrelated. However, reducing risk and vulnerabilities is not the only way to increase resilience in a system. e antici- pated planning on how to respond, recover, and adapt to possible disruptions also plays a vital role in improving the resilience of a system. Resilience from the DOT perspective has three dierent viewpoints: planning, engineering, and operations over two dierent periods: pre-event (risk reductions) and post-event (conse- quence reduction) (AASHTO 2016b). (Adapted from DHS 2010.) Figure B-4. Relationship between risk, resilience, and vulnerability.

Literature Review B-15 3.4 Frameworks for Risk and Resilience Addressing the challenge of integrating resilience into all facets of a transportation agency’s activities requires a framework. Risk and resilience frameworks provide a conceptual structure, logic, and potentially various tools to frame and estimate risk and resilience. Different risk and resilience frameworks have been developed and applied in the transportation sector. One of the most common frameworks developed for risk management is the ISO’s 31000:2009 (International Organization for Standardization 2018). This framework can be applied to any field and is based on steps that should help identify risks and develop a communications struc- ture with stakeholders (see Figure B-5). Another framework developed by the ASME has been implemented in different sectors, including transportation. ASME forged a management process to estimate risk and resilience of various assets from multiple hazards. This process is called RAMCAP Plus (ASME 2009). RAMCAP includes a seven-step framework process that estimates and manages risk and resilience. Figure B-5 shows the RAMCAP Plus framework. Figure B-6 summarizes some common indus- try frameworks and risk and resilience assessment and management standards. Table B-5 summarizes several risk assessment frameworks from around the world. There are no standard frameworks to estimate risk and resilience in the transportation sector. However, many efforts from FHWA and AASHTO have focused on developing frameworks for risk management. To help transportation agencies to incorporate risk management in their TAMPs, the AASHTO Transportation Management Guide: A Focus in Implementation illus- trates a typical risk management framework (see Figure B-7) that can be used for multiple activities in an organization from specific projects to corporate management (AASHTO 2013). In addition, the U.S. DOT’s Volpe National Transportation Systems Center drafted a resil- ience framework in 2013 (Volpe 2013). Volpe’s framework is based on defense in depth, life- cycle cost analysis, and adaptative measures. Figure B-8 presents a diagram of the Volpe resilience framework. Figure B-5. The seven steps of the RAMCAP Plus process.

B-16 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis (Adapted from International Organization for Standardization 2018.) Figure B-6. ISO’s 3100:2009 risk framework. Key: CSA—Canadian Standards Association; BS—British Standard; IRGC—International Risk Governance Council; COSO—Committee of Sponsoring Organizations; AS/NZ—Standards Australia and Standards New Zealand; ISO—International Standards Organization; ASME— American Society of Mechanical Engineers. CSA 1997 BS 6079-3 (2000) IRGC 2004 COSO (2004) AS/NZ4360 (2004) ISO 31000 (2009) ASME RAMCAP Plus (2009) 1. initiation 2. preliminary analysis 3. estimation 4. evaluation 5. control 6. action or monitor 7. communicate 1. context 2. identification 3. analysis 4. evaluation 5. treatment 6. communicate 7. review and update 1. pre- assessment 2. appraisal 3. tolerability judgment 4. risk management 5. communicate 1. environment 2. objectives 3. identification 4. assessment 5. response 6. control 7. communicate 8. monitoring 1. context 2. identification 3. analysis 4. evaluation 5. treatment 6. communicate 7. monitor and review 1. mandate 2. context 3. identification 4. analysis 5. evaluation 6. treatment 7. communicate 8. consult 9. monitor and review 1. asset characteriza- tion 2. threat characteriza- tion 3. consequence analysis 4. vulnerability analysis 5. threat assessment 6. risk and resilience assessment 7. risk and resilience management Table B-5. Example of industry risk and resilience frameworks/standards.

Literature Review B-17 (Source: AASHTO 2013.) Figure B-7. Typical risk management framework. (Source: Volpe 2013.) Figure B-8. Volpe infrastructure resilience framework.

B-18 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis Many other frameworks have been developed in the public and private sectors applicable to risk management. Proctor et al. (2016) provides an overview of some of these frameworks in their final literature review report. A common theme in most of these frameworks is the identification of vulnerabilities, con- sequences, and threats that may cause disruptions to the transportation network or system, along with the title of possible mitigation strategies to reduce the risk and increase resilience. These frameworks or processes provide a tool to stakeholders to help them plan and prioritize investments. In summary, a framework is a system for solving complex problems. Some proposed resil- ience frameworks emphasize adaptability, recovery, and long-term planning, while others focus on transportation’s core mission of providing reliable mobility to all users. 4 Methods of Risk and Resilience Assessment, Metrics, and Performance Indicators As found in literature and practice, there are many definitions for risk and resilience and no standard methodologies or metrics associated with these terms. This section presents an over- view of some assessment methodologies and metrics for risk and resilience and performance indicators used in the industry. 4.1 Risk Assessment Methodologies and Metrics Risk analysis or assessment may be classified into qualitative, quantitative, and a combination of both. These models may vary by degree of complexity. Some use simple methods and others applying more sophisticated probabilistic methods. This next section provides an overview of these methods. 4.1.1 Qualitative Risk Assessment Methods Qualitative risk models are widely used throughout the public and private sectors because of their ease of use and limited data requirements. They can be applied when analyzing strategic goals and related items. Qualitative risk models produce non-numerical estimates of risk using categorical metrics like “low,” “medium,” and “high,” or a scale of discrete numbers, as illustrated in Figure B-9 (FHWA 2017). In addition, other examples of qualitative approaches include risk matrices and weighted index models where the weighted score is calculated for each variable based on an integer scale, (Adapted from FHWA 2017.) Figure B-9. Example qualitative risk matrix.

Literature Review B-19 e.g., from 0 (low) to 4 (high). e integer scores reect the relative signicance of a given vari- able’s value and should be based on expert opinion. Risk matrices are employed across all sectors because of their simplicity. Unlike multi-criterion models that can be implemented with GIS or spreadsheets, risk matrices are usually built with spreadsheet tools, such as Microso Excel. A risk matrix is a two-dimensional relative risk model. e matrix ranks consequences on the horizontal axis and relative likelihood on the vertical axis. Subject matter opinion is relied upon, in many cases, to determine likelihood and consequences. Figure B-10 presents a qualitative risk assessment and metrics (FHWA 2017). Risk matrices are widely used in the transportation sector. Figure B-10 also shows an example of a qualitative risk metric used in transportation risk management (Curtis et al. 2012). An example of risk matrices can be seen in risk registers. Risk registers (see Figure B-11) are widely used by transportation agencies to address and document risks. e risk registers include risk characteristics, the cause, aected objectives, likelihoods, impacts, and nal risk ratings based on these conditions. In 2012, the FHWA published a series of reports on Risk-Based Transportation Asset Management, and Report 3 described a framework for how risk manage- ment principles can be incorporated into TAM plans (FHWA 2012). Report 3 provided guid- ance to shi risk management from a project-level practice to a broader strategic-level practice to cover all aspects of asset management. Using risk management in this way can help to mitigate the impacts of crises through proactive action to improve asset maintenance procedures and policies. As part of a risk management methodology, this report encourages agencies to develop an Asset Risk Register. An Asset Risk Register is a spreadsheet or database that lists an organiza- tion’s identied risks, ranks them based on likelihood and potential impact, and explains how (Source: FHWA 2017.) Figure B-10. Heat map example for risk assessment.

B-20 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis those risks are being treated or mitigated. This register can make strategic decisions to mitigate the most dangerous risks within an agency’s resource limitations and risk tolerance. This meth- odology is primarily qualitative, focusing on identifying risks and ranking risk severity. O’Har et al. (2016) provides a template and instructions for creating a risk register. This tool goes beyond asset-related risks and includes templates for enterprise-level and program-level risk management. Users can use the spreadsheet to manually summarize information about spe- cific risks facing an enterprise or a program within different categories and use the automated features of the tool to populate the risk register and develop a pre-mitigated and post-mitigated heat map of risks, and a risk summary. However, this tool is not a substitute for the executive (Source: FHWA 2012.) Figure B-11. Sample risk register.

Literature Review B-21 leadership, governance, training, and awareness required for successful Enterprise Risk Man- agement implementation. 4.1.2 Quantitative Risk Assessment Methods There are a variety of options states can choose to move forward with risk in their TAMPs, including moving beyond a risk register. Many states are choosing to move beyond risk registers to develop more measurement-based ways to assess risk within their transportation agency. Some of those different processes include quantitative risk assessments (QRA). A formalized and systematic risk analysis approach, a QRA is an essential strategy to understand risk expo- sure. It also provides insight to make cost-effective decisions and manage life-cycle risks (Engineering Safety Consultants n.d.). The QRA process reflects a quantitative representation of risk numerically using models that may be simple or complex and deterministic or probabilistic. Deterministic risk models are based on single values or points and inputs to the equations or model; therefore, these models also generate a single output. An example of the applica- tion of this type of model has been incorporated into the ASME methodology for RAMCAP Plus (ASME 2009). The RAMCAP methodology is an all-hazards risk and resilience manage- ment process for critical infrastructure. Equation 1 expresses how risk is calculated based on the parameters mentioned: ( ) ( )( )= × × (Eq. 1)Risk Threat Vulnerability Consequences where: • Risk is the potential for loss or harm due to the likelihood of an unwanted event and its adverse consequences. When the probability and consequences are expressed as numerical point estimates, the expected risk is computed as the product of those values. • Threat (T) is the likelihood that an adverse event will occur within a specified period, usually 1 year. The event could be anything with the potential to cause the loss of or damage to an asset or population. • Vulnerability (V) is the probability that, given an adverse event, the estimated consequences will ensue. • Consequence (C) is the outcome of an event occurrence, including immediate, short- and long-term, direct, and indirect losses and effects. Loss may include human fatalities and inju- ries, economic damage, and environmental impacts, which can generally be estimated in quantitative terms, and less tangible, non-quantifiable effects, including political ramifica- tions, decreased morale, reductions in operational effectiveness, or military readiness, etc. This approach has been used in multiple sectors, including water and wastewater. The American Water Work Association (AWWA) developed its Risk and Resilience Management of Water and Wastewater System Standard (J-100) based on the RAMCAP framework (AWWA 2010). Since quantitative risk assessments are a reasonably new approach, this method is not widely used in the transportation sector. However, some transportation agencies, including state DOTs, have begun incorporating this approach. Some of the state DOTs that have implemented this approach include Minnesota, Nevada, New York, Utah, Virginia, Washington, and Colorado (Caltrans Division of Research, Innovation and System Information 2015). The Colorado DOT (CDOT) first implemented this methodology to “build back better” after a 2013 major flood event. This work continued with a pilot project sponsored by FHWA to esti- mate the annual risk from multiple threats to highway assets on I-70 (AEM Corporation 2017). Following this pilot, CDOT invested in the development of a Risk and Resilience Procedure

B-22 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis (AEM Corporation 2020) to help guide CDOT staff on the calculation of risk and resilience to multiple highway assets, including culverts, bridges, and roadways from multiple threats such as flood, scour, rockfall, and post-fire debris flow (see Figure B-12). Similarly, Utah DOT (UDOT) was also sponsored by FHWA Extreme Weather and Climate Change Vulnerability Assessment to develop a pilot project which also implemented the RAMCAP framework and methodology to estimate the risk to their highway infrastructure on US-40 from multiple applicable threats. Washington State was an early adopter of the qualitative methods, beginning the develop- ment of a formal risk assessment process in 2002. Nevada DOT implemented its version of a qualitative method in 2008 and New York State DOT in 2009. UDOT started by investigating Washington State DOT’s (WSDOT’s) qualitative methods and then implementing their pro- gram. MnDOT began implementing these practices in 2013 and Virginia in 2011 (Caltrans Division of Research, Innovation and System Information 2015). Minnesota, Nevada, New York, Utah, and Washington have similar qualitative risk assess- ment methods. They start with a workshop to identify and quantify risks. They then utilize a tool to assess the data gathered and conduct a follow-up to track risk throughout a project’s development. In Minnesota, projects identified as “major projects” are subject to quantitative risk analy- sis requiring a consultant-led analysis workshop to identify project risks. MnDOT utilizes the @RISK tool by Palisade to identify probabilities and risks associated with alternatives (Caltrans Figure B-12. CDOT risk assessment procedure.

Literature Review B-23 Division of Research, Innovation and System Information 2015). Most projects subjected to Nevada’s comprehensive quantitative risk analysis are large projects that start with a facilitated workshop. They utilize a risk management tool that generates probabilistic modeling and risk registers. Utah uses qualitative and quantitative tools to assess risk during project development, typically as early as possible. For many states, like Washington and Virginia, one of the key out- puts of their qualitative risk assessment is a risk register. While states have noted challenges to getting their qualitative risk assessment programs off the ground, there are benefits resulting from using this new method. MnDOT has found cost estimates are closer to or under budget, UDOT has benefited from the increased communica- tion due to the workshops, and WSDOT believes it better assists in addressing issues that arise in the field (Caltrans Division of Research, Innovation, and System Information 2015). Probabilistic models attempt to account for uncertainty by employing random variables with their respective probability distributions to produce a set of possible outcomes for every risk calculation. Deterministic models are entirely determined by their parameter values and initial condi- tions. However, natural phenomena are stochastic in nature. Probabilistic models share greater fidelity with biological processes, capture uncertainty, and quantify risk using values from the physical world. Probabilistic approaches include Bayesian Belief Networks, Markov chains, and Monte Carlo simulation. These models are recommended when there is a certain degree of uncertainty on the parameters used to calculate vulnerability or risk. The European Union developed an example of a probabilistic risk assessment model named INFRARISK to evaluate risks to multiple infrastructure networks from multiple threats (INFRARISK n.d.). This model uses Monte Carlo simulations, which use random sampling for a given variable to generate a set of possible numerical results. For example, INFRARISK uses Monte Carlo simulation to estimate the probability of ground motions (earthquakes). Examples of transportation agencies in the United States employing probabilistic methods include (1) the partnership between Arizona Water Science Center (AZWSC) and the Arizona DOT (ADOT), and (2) the Mechanistic Empirical Pavement Design Guide (MEPDG). The partnership between AZWSC and ADOT has resulted in improved stormwater manage- ment. This includes data collection and modeling at key water crossings to provide data on flood magnitude and measure changes in cross-sectional area to refine the accuracy of hydrologic models and how channel conditions affect infrastructure. This partnership between AZWSC and ADOT led to a rapid installation of gauges and sensors to monitor hydrologic parameters, including stormwater runoff, peak flow, velocity, and topo- graphic surveys to improve bank stabilization projects and bridge and culvert design. A GIS Resilience Database was created to identify areas at risk, using data collection and new tech- nology for proactively addressing asset management issues. As part of its 2019 resilience pilot project (ADOT 2019), ADOT developed a quantitative probabilistic methodology for bridge asset class considering resilience in infrastructure life-cycle management. The MEPDG was used to develop a probabilistic pavement fragility model for generating fragility functions, which integrates flood hazards, pavement structure, pavement performance, and damage states (Lu et al. n.d.). NCHRP Report 602 documents the calibration and validation of the Enhanced Integrated Climatic Model (EICM), a one-dimensional coupled heat and mois- ture flow program intended to analyze pavement-soil systems for Pavement Design (Zapata and Houston 2008).

B-24 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis 4.1.3 Hybrid Risk Assessments Methods Hybrid methods are a combination of qualitative and quantitative methods. The World Health Organization (WHO) (2009) explains that semiquantitative risk assessment bridges the gap between purely descriptive measures of risk and numerical measures using a scoring system. A common semiquantitative method is the multi-criterion weighted index model. Multi- criterion, weighted index models are frequently used in GIS to map hazard susceptibility or to identify sites most suitable for land use. A multi-criterion risk model calculates risk as to the sum of the products of individually ranked criteria (see Equation 2) (Stanujkic and Zavadskas 2015). ∑= × = (Eq. 2) 1 R W Ci i i n where: R = total relative risk. Wi = weighting. Ci = criterion. This type of model falls under the umbrella of relative risk models because the selection of criteria and the ranking and weight are subjectively selected. There are different ways to generate the weighting criterion in these models, either expert opinion or more sophisticated method- ologies. For example, if the slope is three times as important as geology (rock type), the slope weighting criteria may be assigned a weighting value of 3, while geology is given a weighting value of 1. An example of a hybrid model is the FHWA VAAF where vulnerability is scored as a function of multiple components (exposure, sensitivity, and adaptive capacity) (Filosa et al. 2017). Users can adjust the scoring by assigning weights to those parameters. Two tools have been implemented to help transportation agencies put the VAAF into prac- tice, the Vulnerability Assessment Scoring Tool (VAST) and the Coupled Model Intercompar- ing Project (CMIP) Climate Data Processing Tool. Both tools have been used extensively by state DOTs and MPOs in FHWA-sponsored risk and resilience pilot studies. VAST is a spreadsheet tool that guides planners to conduct a vulnerability screen of their assets (Adaptation Clearing House 2015). This tool uses an indicator-based approach to deter- mine how assets will respond to climate stressors and results in a weighted vulnerability score. There are three components of vulnerability measured: • Exposure: Whether an asset will experience a given stressor. • Sensitivity: Whether an asset will be damaged or disrupted by a given stressor. The worse an asset’s condition, the more likely it will be damaged if exposed. • Adaptive capacity: How well the system can cope with damage or disruption to specific assets. This is influenced by data such as usage statistics or traffic volumes. The outputs of this tool can be used as a component of the calculation of vulnerability used in either a probabilistic or deterministic risk assessment model. The CMIP Climate Data Processing Tool 2.1 is an online web application that assists planners in downloading data from the CMIP databases and processing that data into statistics relevant to transportation planners (FHWA 2020a). Users start by accessing the CMIP database and following the tool’s instructions to download data on specific grids that encompass the loca- tions they are interested in examining (Downscaled CMIP3 and CMIP5 Climate and Hydrol- ogy Projections, n.d.). The tool processes the downloaded climate data to produce a variety of

Literature Review B-25 climate-related statistics related to events like heat, cold, and precipitation that can be used to estimate the impact of dierent risks. Another example of a risk assessment model that uses assigned scores to multiple parameters was developed by Pennsylvania DOT (PennDOT) as part of an extreme weather vulnerability study and climate change adaptation initiatives. Figure B-13 shows the scoring methodology and risk equation developed for this project (PennDOT 2017). (Source: PennDOT 2017.) Figure B-13. PennDOT risk assessment model and cumulative scoring formula.

B-26 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis 4.1.4 Risk Metrics Since there is no standard risk assessment methodology, there are no standard risk metrics, especially in the transportation sector. In the field of economics, Holton (2014) distinguished between risk measures—“the opera- tion that assigns a value to a risk” and risk metrics—“the attribute of risk that is being measured” and stated that risk metrics commonly take one of the following three forms: those that quantify exposure, uncertainty, or exposure and uncertainty in some combined manner. Depending on how risk is determined, qualitative, quantitative, or both, the metrics used to quantify risk may vary. Table B-6 describes some examples of risk metrics for measuring eco- nomic and population risk (Westen et al. 2011). A common example of qualitative risk metrics is presented in Figure B-14, where risk is estimated using risk matrices and reported ranging from “Low” to “High.” Figure B-14 and Figure B-15 present examples of risk matrices for a 2015 FHWA pilot project to estimate risk for the Central Texas regional transportation infrastructure (Cambridge Systematics and ICF International 2015). As an example of a quantitative risk metric, the ASME’s RAMCAP Plus methodology for risk assessment measures risk as to the product of threat likelihood, vulnerability, and consequences Type Sub-Type Principle Q ua lit at iv e Qualitative Based on relative risk classes categorized by expert judgment. Risk classes: High, Moderate, and Low Semi- quantitative Based on relative ranking and weights assignments by criteria. Risk index: ranked values (0-1, 0-10 or 0-100). (dimensionless) Q ua nt ita tiv e Probability Probabilistic values (0-1) for having a predefined loss over a particular time period Economic Risk Quantification of the expected losses in monetary values over a specific period of time Probable Maximum Loss (PML) Probable Maximum Loss (PML) The largest loss believed to be possible in a defined return period, such as 1 in 100 years, or 1 in 250 years. Average Annual Loss (AAL) Expected loss per year when averaged over a very long period (e.g., 1,000 years). Computationally, AAL is the summation of products of event losses and event occurrence probabilities for all stochastic events in a loss model. Loss Exceedance Curve (LEC) Risk curve plotting the consequences (losses) against the probability for many different events with different return periods. Quantification of the risk to the population Population risk Individual risk The risk of fatality or injury to any identifiable (named) individual who lives within the zone impacted by a hazard; or follows a particular pattern of life that might subject him or her to the consequences of a hazard. Societal risk The risk of multiple fatalities or injuries in society as a whole. Society carries the burden of a hazard that causes deaths, injury, financial, environmental, and other losses. Table B-6. Different ways of expressing risk.

Literature Review B-27 (Source: Cambridge Systematics and ICF International 2015; Winter et al. 2014.) Figure B-14. Climate change risk matrix, 2015 Central Texas extreme weather and climate vulnerability study. (Source: Cambridge Systematics and ICF International 2015.) Figure B-15. Risk rating summary for the 2015 Central Texas extreme weather and climate change vulnerability assessment of regional transportation infrastructure.

B-28 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis (ASME 2009). Since these parameters can be estimated as probabilities (or percentages) for exposure and threat likelihood and in dollars for consequences, the product provides a risk result in dollars. CDOT and UDOT adopted the RAMCAP risk calculation methodology to estimate the risk to their highway infrastructure from multiple threats and use risk metrics represented in dollars per year [e.g., annual risk for culverts from flooding ($/year)]. Moreover, the Virginia Transportation Research Council identified four risk metrics for describing the consequences of storm surge and high winds (Ferguson 2002): • Ratio of repair cost to reconstruction cost • Repair cost • Time to recover • Cost to industry In the rail sector, the Federal Railroad Administration (FRA) describes risk metrics as the “primary quantitative inputs for quantitative risk analysis related to hazmat transportation safety.” The risk metrics (see Table B-7) are conditional probabilities of accident likelihood used as inputs to a fault tree analysis (Federal Railroad Administration 2015). 4.1.5 Risk-Based Performance Measures Different performance measures and targets are used in transportation. Transportation agencies use these to measure how their agencies perform over time and if they are achieving their set targets. Performance measures (goals or key performance indicators (KPI)) must be distinguished from performance metrics. Performance metrics are the numbers or qualitative characteristics within a performance measure that help track performance and progress. For example, U.S. DOT’s Fiscal Year 2017 Annual Performance Report lists the following per- formance measure under the goal of roadway safety: “highway fatality rate per 100 million vehicle miles traveled with the target number of fatalities being 1.02 (U.S. DOT n.d.). This performance measure contains two metrics: the number of fatalities and the millions of miles traveled. MAP-21 requires transportation agencies to develop a risk-based performance manage- ment asset management plan to be reviewed every 4 years. The stated purpose of MAP-21 is the “establishment of a performance- and outcome-based program (FHWA 2013). Table B-8 lists seven goal areas and their respective national goals for the transportation sector, as cited in §1203: 23 USC 150(b). Risk Metric Function of Train accident frequency Accident cause, FRA Track Class, Railroad Type (Class I or non-Class I) Car derailment probability Train speed, train length, loading The conditional probability of a car containing hazmat Relative volumes of freight and hazmat shipments on route being analyzed Conditional probability of release from derailed car Tank car design/construction, train speed, accident cause Conditional probability of harm from release Hazmat type, emergency response action (Source: Federal Railroad Administration 2015.) Table B-7. FRA risk metrics.

Literature Review B-29 Further, §1203; 23 USC 150(c) lists the following performance measures: • Pavement condition on the Interstate System and on the remainder of the National Highway System (NHS) • Performance of the Interstate System and the remainder of the NHS • Bridge condition on the NHS • Fatalities and serious injuries—both number and rate per vehicle mile traveled—on all public roads • Traffic congestion • On-road mobile source emissions • Freight movements on the Interstate System Along with the requirements from MAP-21, AASHTO also provides a series of performance measures in their Transportation Asset Management Guide: A Focus on Implementation. The performance measures included in the guide are as follows: • Condition • Life-cycle cost • Safety • Mobility • Reliability • Customer measures • Externalities • Risk To track success in meeting these goals, state DOTs establish their performance measures, typically outlined in the TAMP. Here are some examples: • WSDOT tracks percentage of pavement that falls within a given condition state, for example, the rate of pavement (all state roads) in fair or better condition with a target of 85% (Miller 2018). Goal area National goal Safety To achieve a significant reduction in traffic fatalities and serious injuries on all public roads Infrastructure condition To maintain the highway infrastructure asset system in a state of good repair Congestion reduction To achieve a significant reduction in congestion on the National Highway System System reliability To improve the efficiency of the surface transportation system Freight movement and economic vitality To improve the national freight network, strengthen the ability of rural communities to access national and international trade markets, and support regional economic development Environment sustainability To enhance the performance of the transportation system while protecting and enhancing the natural environment Reduced project delivery delays To reduce project cost, promote jobs and the economy, and expedite the movement of people and goods by accelerating project completion through eliminating delays in the project development and delivery process, including regulatory burdens, and improving agencies’ work practices Table B-8. MAP-21 goal areas and national goals.

B-30 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis • MnDOT’s performance measures for operational excellence include the percentage share of the Interstate system with poor ride quality in the travel lane, with a target of no more than 2% (MnDOT 2018). • UDOT’s performance measure for signal systems is that 95% of the systems are rated as average or in good condition (UDOT 2019). • ADOT’s performance target for NHS bridges is 52% of bridges classified as in good condition (2- and 4-year targets) (Anderson 2019). Flannery et al. (2019) found that information needed to develop resilience metrics and assess- ment methods was a “pressing need” for state DOTs. After reviewing the planning documents of 101 MPOs and 50 state DOTs, Dix et al. (2018) found that only five state DOTs established resilience-related performance measures compared with 19 MPOs. In addition, state DOT per- formance measures tended to be tied to safety and security, environmental stewardship, or system preservation (asset management) and did not address natural hazards or extreme events. In contrast, 19 MPOs included performance measures directly related to promoting resilience against climate change and extreme events. Table B-9 gives example performance measures for adaptation to climate change for nine MPOs. Asam et al. (2015) adapted a list of sample operations and performance measures from FHWA’s desk reference, Advancing Metropolitan Planning for Operations, for possible use as objectives and measures for improving resilience to climate change (FHWA 2010) (see Table B-10). Given that resilience has an inverse relationship with risk (see the section on the relationship between risk and resilience), i.e., increasing resilience decreases risk, the performance measures listed in Table B-9 and Table B-10 could serve as examples for risk-based performance metrics. The transportation sector has identified the need to incorporate risk and resilience into performance measures (risk-based performance measures); however, there is still no standard process or measure. 4.2 Resilience Assessment Methodologies and Metrics Federal initiatives such as MAP-21 and the FAST Act compel transportation agencies to assess the resilience of their systems. In a frequently cited paper, Bruneau and Reinhorn (2006) outlined the four “Rs” of resilience: robustness, redundancy, resourcefulness, and rapidity. • Robustness. “Strength, or the ability of elements, systems, and other measures of analysis to withstand a given level of stress or demand without suffering degradation or loss of function.” • Redundancy. “The extent to which elements, systems, or other measures of analysis exist that are substitutable, i.e., capable of satisfying functional requirements in the event of a disrup- tion, degradation, or loss of functionality.” • Resourcefulness. “The capacity to identify problems, establish priorities, and mobilize resources when conditions exist that threaten to disrupt some element, system, or other measures of analysis.” • Rapidity. “The capacity to meet priorities and achieve goals in a timely manner to contain losses, recover functionality and avoid future disruption.” Note that these metrics align with FHWA’s sensitivity and adaptive capacity concepts. Other suggested metrics for the four “R’s” relevant to transportation can be found in Table B-11 (Parkany and Ogunye 2016). An exhaustive study of transportation metrics was completed by Sun et al. (2018), describ- ing three categories of metrics: topological, traffic-related, and functional (see Table B-12) (Sun et al. 2018). Topological might be regarded as measures of redundancy, one of the four “R’s.” Traffic-related metrics assess network performance by measuring the amount of traffic flowing through the network.

Literature Review B-31 MPO Goal or Objective Performance Measure(s) or Targets Cape Cod MPO (MA) (Cape Cod MPO 2015) Improve the transportation system's resiliency to the effects of sea-level rise. Evaluate potential impacts of sea- level rise for all TIP projects during the 25% design review, and adjustments to projects are made as warranted. Improve stormwater management and treatment in transportation improvement projects. Provide improved stormwater management and treatment to 50% of TIP projects outside of sensitive areas and 100% of TIP projects within sensitive areas. Hillsborough County MPO (Tampa, FL) (Hillsborough County MPO 2016) Increase the security and resiliency of the multimodal transportation system. Protect low-lying major roads from storm surge and flooding. Maintain stormwater drainage programs. Merrimack Valley MPO (Haverhill, MA) (Merrimack Valley MPO 2015) Use adaptive planning for climate change. Number of coastal communities with adaptation plans. Miami-Dade MPO (FL) (Miami-Dade MPO 2014) Reduce the vulnerability and increase the resiliency of critical infrastructure to the impacts of climate trends and events. Number of highway lane and centerline miles within the 100-year floodplain. Mid-Region COG (Albuquerque, NM) (Mid-Region COG 2015) Environmental resilience: Prepare for climate uncertainties. Development in high flood risk areas: Employment and housing in FEMA 100-Year floodplains. Development in forest fire risk areas: Employment and housing in wildland-urban intermix areas. Northern Middlesex MPO (Lowell, MA) (Northern Middlesex MPO n.d.) Protect critical infrastructure from the effects of climate change and address stormwater runoff and flooding concerns. Number of stormwater improvement projects implemented by local communities and MassDOT. Palm Beach MPO (FL) (Palm Beach Metropolitan Planning Organization 2014) Provide an efficient and reliable vehicular transportation system. Increase the percentage of facilities that accommodate 2-foot sea-level rise; the performance target is 90% for the strategic intermodal system network in 2025. Regional Planning Commission (New Orleans, LA) (Regional Planning Commission 2015) Environmental sustainability: implement projects that consider the impacts of climate change and natural hazard mitigation. Number of projects that raise the roadway grade or increase resilience against climate change or natural disasters through other means (tracked annually). Tri-County Regional Planning Commission (Peoria, IL) (Tri-County Regional Planning Commission 2015) Provide an efficient and resilient transportation system. Ensure 95% of all roadways have a volume-capacity ratio less than one by 2020. Reduce the percentage of roadways in “poor” or “fair” condition. Reduce the percentage of roadways in “critical backlog.” Reduce commute times by 2.5% by 2025. Table B-9. MPO performance measures.

B-32 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis Category Operations Objectives Performance Measures Clearance Time (Weather- Related Debris) Reduce average time to complete clearing (mode, hierarchy of facilities, or subarea of region) of weather- related debris after weather impact by X% in Y years. Reduce average time to complete clearing (interstates, freeways, expressways, all roads, main tracks, and main sidewalks) of weather-related debris after weather impact by X% in Y years. Average time to clear selected surface transportation facilities of weather- related debris after weather impact. Detours for Impacted Roadways Increase by X% the number of significant travel routes covered by weather-related diversion plans by year Y. Increase the percentage of agencies that have adopted multi- agency weather- related transportation operation plans and that are involved in transportation operations during weather events to X% by year Y. Percentage of significant travel routes covered by weather-related diversion plans. Percentage of agencies involved in transportation operations during weather events that have adopted multi-agency, weather- related transportation operations plan. Disseminating Information Reduce time to alert travelers of travel weather impacts [using variable message signs, 511, Road Weather Information Systems (RWIS), public information broadcasts, the agency's website, Web 2.0 technologies, etc.] by X (time period or percent) in Y years. Time from beginning of weather event to posting of traveler information on variable message signs, 511, RWIS, public information broadcasts, etc. Time from beginning of weather event to posting of traveler information on agency website. Increase the percentage of major road network (or transit network Percentage of major road (transit or bicycle) network Road Weather Information System Coverage or regional bicycle network) covered by weather sensors or an RWIS by X% in Y years as defined by an RWIS station within Z miles. within Z miles of an RWIS station. Signal Timing Plans Special timing plans are available for use during inclement weather conditions for X miles of arterials in the region by year Y. Number of miles of arterials that have at least one special timing plan for inclement weather events. Table B-10. Sample operations objectives and performance measures.

Literature Review B-33 Class Metric Description Topological Centrality Measure of influence of a node in a network. Weighted centrality Same as centrality except links between nodes are weighted. Weights are typically based on link length, travel time, or travel distance. Traffic Travel time Travel time used to measure travel delay during and post-event. Throughput The total sum of flows of shipments/passengers between origins and destinations. Congestion index A measure of travel delays due to a disruption, e.g., link delay to acceptable travel time. Functionality Resilience triangle A graph of functionality recovery starting from the extreme event until full recovery. Resilience index Based on the resilience triangle, a measure of functionality over time. Capacity Absorptive Ability to absorb perturbations from an event. Adaptive Ability of system to gradually adapt itself to a disruption. Coping Ability to respond to and recover from events. Restorative Ability to bounce back to the original performance level or better. Table B-12. Resilience metrics. Resilience Pillar Indicator Robustness Hours of congestion Spatial extent of congestion Travel time index Optimal spare capacity Pavement condition Weather impact Volume of congestion Redundancy Distance to alternative routes Percentage of corridor(s) with alternate routes Available capacity on alternative routes Congestion on alternative routes Graph theory connectivity score Transit alternatives Adjacent park-and-ride lots Resourcefulness Safety service patrol Average incident duration Availability of special transportation funding Message signs Weather stations The use of alternative routes Construction projects Weather mitigation capability Rapidity Regain time for top 5% incident Average construction project duration Table B-11. Resilience metrics for transportation.

B-34 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis Functionality metrics involve a before and after comparison of network performance and evaluate the network’s ability to resume normal functionality after a disruption. For example, in the resiliency triangle (see Figure B-16), Bruneau et al. (2003) show that a system is fully func- tional before an event, functionality drops dramatically immediately following an event, and then, there is a gradual return to full functionality. Other metrics of resilience have been developed and used in the transportation sector. An exam- ple includes a qualitative metric developed by AEM Corporation called Level of Resilience (LOR) Index, which is based on criticality scores of a system or network (low, medium, or high) and the cumulative annual risk of the system from multiple threats estimated in dollars per year (see Fig- ure B-17). Figure B-18 shows an example of the application of this metric on CDOT’s I-70 Pilot, where the LOR index varies from A through E, where LOR A means the system or network has a “Very High” resilience and LOR E means it has a “Very Low” Resilience (AEM Corporation 2017). UDOT also developed a resilience assessment metric as part of its risk management process. Noting that resilience is “inversely proportional to risk and criticality,” UDOT formulated the fol- lowing equation (see Equation 3) (Alder et al. 2020). UDOT’s resilience metric also demonstrates the relationship between risk and resilience. ( )= ×1 (Eq. 3)Resilience risk criticality (Adapted from Bruneau et al. 2003.) Figure B-16. The resiliency triangle. (Source: AEM Corporation 2017.) Figure B-17. Level of resilience (LOR) matrix, from CDOT I-70 Corridor Risk and Resilience Pilot.

Literature Review B-35 In summary, researchers and practitioners have developed several metrics to estimate the risk and resilience of transportation networks. However, there is still no agreement on assessment methodology, standard metrics, or performance indicators. The National Academies of Sci- ences, Engineering, and Medicine recently established a full committee named Transportation Resilience Metrics to identify and examine these metrics (https://www.nationalacademies.org/ our-work/transportation-resilience-metrics). 5 Major Components of a Risk Assessment Methodology This section discusses the major components of a risk assessment methodology: asset characterization (criticality), threat assessment, vulnerability assessment, and consequence modeling. 5.1 Methods to Identify Critical Transportation Assets Determining asset criticality is one of the first steps in risk and resilience analysis, “Asset Characterization.” Criticality is a measure of the importance of an asset to the system. ICF International’s report for U.S. DOT, Assessing Criticality in Transportation Planning, describes criticality in the context of climate change vulnerability assessments as “. . . those assets of ‘greatest importance,’ such as assets that are of economic importance, provide access to healthcare facilities, serve as emergency evacuation routes, provide social connectivity, have cultural significance, or support other core values” (ICF International 2014). (Source: AEM Corporation 2017.) Figure B-18. Example of application of LOR index on CDOT’s I-70 Risk and Resilience Pilot.

B-36 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis The ICF International report (2014) describes three approaches for assessing criticality: (1) desk review, (2) stakeholder elicitation, and (3) hybrid approach. • Desk review. This approach is essentially an index method that quantitatively compares and ranks assets based on a set of criteria, such as AADT, functional class, network redundancy, etc. (ICF International 2014; Hampton Roads Transportation Planning District Commis- sion 2014). Example risk and resilience studies that employed this approach include Virginia DOT’s Hampton Roads pilot, which used traffic volume, proximity to a hurricane evacuation route, location on a maintenance route, and relative elevation compared with mean sea level, and New Jersey’s pilot, which based its criticality assessment on the importance of destina- tions (population density and jobs), traffic volume, and the emergency function of routes (Cambridge Systematics 2011; ICF International 2014). • Stakeholder elicitation. This approach assembles stakeholders and subject matter experts to participate in one or more workshops to elicit expert opinions on which assets are critical. For example, for WSDOT’s pilot study, Operations and Maintenance (O&M) personnel and engineers participated in workshops to rank asset criticality on a scale of 1 to 10, based on such factors as AADT, functional class, availability of alternate routes, and whether it is part of the NHS (ICF International 2014; WSDOT 2011). • Hybrid approach. This approach combines the desktop review with stakeholder elicitation. Typically, analysts begin by ranking assets based on various criteria and then solicit expert opinions from SMEs and stakeholders to get feedback on the selection of criteria, weightings, rankings, etc. (ICF International 2014). Table B-13 is an example of a weighted, multi-criterion index model. The criteria include AADT (non-truck), freight (truck AADT), functional clas- sification, and tourism revenue, binned into 5 quantiles where each quantile is assigned an index value from 1 to 5. The criticality score for a given transportation facility is the sum of the indices for each criterion. Different approaches to identify asset criticality can be developed and have been used and tailored to different agency needs. 5.2 Threat Assessment and Incorporation of Uncertainty Another important step of risk and resilience analysis is threat assessment. The elements of threat assessment include threat likelihood (probability of occurrence), magnitude or severity, and extent. The first step is to identify the threats relevant to a given study area. The next step is threat modeling to determine the frequency and magnitude of the threats that have been identified. This section addresses the methodologies and data needs for completing a threat assessment. Table B-13. Example multi-criterion criticality table.

Literature Review B-37 5.2.1 Threat Identification Identifying the threats of concern to a community can be accomplished by reviewing histori- cal records, disaster declarations, hazard maps, and subject matter opinions. Although by no means exhaustive, Table B-14 provides examples of natural, technological, and human-caused threats and hazards that have been identified by many states (DHS 2012). Natural threats result from earth system processes, such as floods, earthquakes, and landslides. Technological hazards are due to accidents or the failure of human-made systems, such as power failures or train derailments. Finally, human-caused threats include intentional acts, such as sabotage and cyberattacks. Naturally, the task is to determine which among these is relevant to the jurisdiction of concern. This task can be accomplished with the help of hazard maps. The relevancy of a hazard to a study area is dependent on geography, climate, population density, and proximity to infrastructure. Using GIS to overlay hazard maps, such as U.S. Geo- logical Survey (USGS) landslide inventories (USGS n.d.a) (see Figure B-19), FEMA National Flood Hazard Layer (NFHL) (FEMA 2021) (see Figure B-20), and NOAA’s historical tornado track database (NOAA 2022) (see Figure B-21), over infrastructure is an effective way to identify relevant threats and isolate exposed assets. The spatial intersection of hazard maps and assets helps define the threat-asset pairs that warrant further analysis. An example of a threat-asset pair matrix is presented in Figure B-22. Threats are listed in the left-hand column, and assets are listed in the top row. Relevant threat-asset pairs are symbolized in orange (lighter shading). The color purple (darker shading) indicates no threat (Institute for Trade and Transportation Studies 2020). 5.2.2 Data Sources to Support Threat Modeling Spatial data in the form of digital maps superimposed over study area boundaries and asset inventory identify the hazards that are relevant to a given study area and identify assets at risk. There is currently a plethora of publicly available spatial data that can be downloaded as a shapefile, KMZ file, GeoJSON, spatial database, or accessed online through a REST service. Key national sources include the U.S. Geological Survey, NOAA, U.S. DHS, and FEMA. In addition, Natural Threats or Hazards Technological Threats or Hazards Human-caused Threats or Hazards Resulting from acts of nature Involves accidents or the failures of systems and structures Caused by the intentional actions of an adversary Flood Earthquake Avalanche Debris flow Wildfire Pandemic Drought Epidemic Hurricane Landslide Tornado Tsunami Volcanic eruption Winter storm Extreme temperature Airplane crash Dam/levee failure Hazardous materials release Power failure Radiological release Train derailment Urban conflagration Bridge strikes Civil disturbance Cyber incidents Sabotage Terrorist acts • • • • • • • • • • • • • • • • • • • • • • • • • • • Table B-14. Examples of threats and hazards.

B-38 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis (Source: USGS n.d.a.) Figure B-19. Landslide inventory, west of Santa Barbara, California. many transportation agencies have invested in developing the spatial data inventories curated by state, county, and local agencies. Table B-15 lists examples of spatial data for modeling hazards, extreme weather, and climate change. These are all free and from public sources. It is important to note that hazard and climate and environmental data may also be available in tabular formats, such as Excel spreadsheets or delimited text files. Table B-16 gives three examples. In addition to the natural hazard and climate-related data sources described, the data sources listed in Table B-17 and Table B-18 may assist in modeling terrorism and cyberattacks, respec- tively. Texas A&M University library hosts a website with links to data sources on terrorism (see Table B-17). Souppaya and Scarfone (2016) discuss modeling cyber threats in detail, while Hug et al. (2017) researched threats to intelligent transportation systems (ITS). Table B-18 lists sources of information relevant to cyber threats. 5.2.3 Types of Threat Modeling This section will cover approaches to threat modeling. Threat modeling encompasses the following aspects: threat extent (spatial boundaries), frequency (likelihood of occurrence), and intensity (magnitude).

Literature Review B-39 (Source: FEMA 2021.) Figure B-20. FEMA NFHL, Port Arthur, Texas. Methodologies for estimating extent depends on the type of threat, (i.e., geotechnical, seismic, or hydrological). Free-ware tools, such as the Flow-R and Q-Proto, a QGIS plugin, and com- mercial tools such as RocScience’s Rockfall, are software solutions for modeling geotechnical events, such as landslides, debris flows, and rockfall. Typical inputs include digital elevation data, slope angle, ground cover, debris volume, and location of the debris source. Figure B-23 demonstrates the use of Flow-R to model the runout zone of a rock avalanche (Oppikoger et al. 2016). The relative probability of propagation is symbolized with a yellow-to-red color ramp. The extent of a seismic threat is determined by attenuation curves. Attenuation curves calcu- late the shaking parameter (e.g., peak ground acceleration, peak ground velocity, etc.) in terms of distance from the earthquake source [i.e., ground shaking is attenuated with distance from the source (FEMA 2020a)]. Shake maps (maps of ground shaking) reveal both the extent and intensity of a seismic event. For example, the shake map for the 1971 San Fernando earthquake (see Figure B-24) symbolizes the region of greatest intensity in red, closest to the earthquake epicenter, while the intensity diminishes with distance from the epicenter, symbolized with a color shift from orange, to yellow, to green (USGS 1971). It is important to note that shake maps illustrate both extent and intensity.

B-40 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis (Source: NOAA 2022.) Figure B-21. Tornado track through Dallas, Texas, October 20, 2019. Note: The orange (lighter) shading indicates all the possible threats for all the assets; the purple (darker) shading indicates the threat-asset combinations selected for analysis. Figure B-22. Example threat-asset matrix for transportation.

Literature Review B-41 Hazard Data Description Source Flood DFIRM Digital flood insurance rate (DFIR) map with boundaries for 100-/500- year floodplain (vector polygon) FEMA: https://msc.fema.gov/ portal/home Peak Flows Watershed basin characteristics and peak flow for 1.25- to 500-year events for ungauged sites, downloadable as vector USGS StreamStats: https://streamstats.usg s.gov/ss/ polygon, CSV, or PDF report. Stream Network ESRI Geodatabase with polyline vector feature classes USGS NHD Plus: https://www.usgs.gov/ national- hydrography/nhdplus- high-resolution NOAA Atlas 14 Rainfall intensity and accumulation by return period in table form NOAA: https://hdsc.nws.noaa. gov/hdsc/pfds/ Real-Time Weather Data Current and forecast weather data OpenWeather: https://openweatherm ap.org/api Earthquake Shake Maps PGA, PGV, spectral acceleration 1-second and 0.3-second periods (HAZUS-ready shapefiles) USGS Earthquake Hazards Program Quaternary Faults Quaternary fault and fold database for the United States, download as KML or polyline shapefile USGS Faults: https://www.usgs.gov /programs/earthquake -hazards/faults?qt- science_support_page _related_con= Historic Epicenters Included are earthquakes located in the United States and some that occurred in adjacent portions of Canada and Mexico. The main sources for the data are Seismicity of the United States, 1568-1989, and the Preliminary Determination of Epicenters for 1990 to August 2009. Homeland Infrastructure Foundation-Level Data Real-time Feed (GeoJSON) Earthquake event data is updated every minute. Can access Significant Earthquakes, M4.5+, etc.) USGS Earthquake Feed Landslide U.S. Landslide Inventory Mostly polygon vector representations of landslide extent points where extent has not been mapped USGS Susceptibility This map layer, utilizing data from the USGS, delineates areas in the conterminous United States where large numbers of landslides have occurred and areas that are susceptible to landslides. USGS Landslide Susceptibility Frequency- Precipitation Estimate the annual frequency of landslides triggered by precipitations. Global Risk Data Platform Table B-15. Examples of special data for hazard modeling. (continued on next page)

B-42 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis Hazard Data Description Source Frequency Earthquake Estimate the annual frequency of landslides triggered by earthquakes. Global Risk Data Platform Elevation/Slope 30-m to 1-m cell resolution digital elevation grids, Lidar USGS – The National Map Soil Various soil characteristics, depth to water, depth to bedrock (Microsoft access database) USDA Web Soil Survey Post Fire Debris Flow U.S. Landslide Inventory Mostly polygon vector representations of landslide extent points where extent has not been mapped USGS Wildfire Fire Perimeters U.S. Historical fire perimeters from 2000 to 2018. Perimeters were originally downloaded from the Geospatial Multi- Agency Coordination (GeoMAC) by USGS. ArcGIS Online Burn Probability The Mean Fire Return Interval (MFRI) layer quantifies the average period between fires under Landfire the presumed historical fire regime. Debris Flow After Fire Debris flow runout zone (vector polygon) with estimated debris volumes and probabilities USGS Post-Fire Debris Flow Assessments Tornado Historical tornado tracks (1950 to 2019) Tornado tracks are available as paths or initial points. NOAA Storm Prediction Center (SPC) Hail Historical hailstorms (1950 to 2019) Hailstorms are available as paths or points. NOAA SPC Wind Historical strong winds (1950 to 2019) Windstorms are available as paths or points. NOAA SPC Hurricane Advisory wind field and forecast wind radii, wind speed probabilities National Hurricane Center data is available as shapefiles and KMZ. National Hurricane Center Storm Surge Hazard Map Gridded storm surge maps were developed using SLOSH Storm Surge modeling. Latest version is 2018. NOAA Sea-Level Rise Depth grid 1-ft to 6-ft sea-level rise inundation extent. Gridded spatial data downloadable by state. Note: the files are large. NOAA Climate Change Downscaled CMIP5 Download LOCA CMIP5 data for multiple time periods and up to 20 Global Climate Models. This data can be processed for transportation usage with Bureau of Land Reclamation the FHWA online CMIP5 tool. Table B-15. (Continued).

Literature Review B-43 Hazard Data Description Source Climate Change CMIP5 downscaled climate data (n.d.) Downloadable in CSV or netCDF Format. Bureau of Land Reclamation Climate Extremes “The U.S. Climate Extremes Index (CEI) is the arithmetic average of the following five or six indicators of the percentage of the conterminous U.S.” Downloadable as an Excel spreadsheet. NOAA: https://www.ncei.n oaa.gov/access/mon itoring/cei/graph Hydrological Peak streamflow for gauged streams. Downloadable as XML file or tab-delimited spreadsheet USGS Table B-16. Example tabular data sources. Data Source Comments Iterate Dataset (Duke University Libraries 2018) “A textual chronology of international terrorism which employs inter alia an exhaustive search of major media including information obtained from interviews with government officials, scholars, and former hostages/others involved in international terrorist incidents.” Global Terrorism Database (University of Maryland 2020) “Information on terrorist events around the world from 1970 through 2013 includes systematic data on domestic and international terrorist incidents that have occurred during this time period and now includes more than 125,000 cases.” Suicide Attack Database (1974-2016) (Princeton University 2016) “Data on all suicide attacks from 1982 through November 2014 including information about the location of attacks, the target type, the weapon used, and systematic information on the demographic and general biographical characteristics of suicide attackers.” Currently offline. Terrorism and Extremist Violence in the United States (TEVUS) database (LaFree et al. 2019) “Integrates existing and new open-source data sets to facilitate more robust and sophisticated analyses of violent extremists' behaviors, operations, and activities within the United States.” American Terrorism Study (1980-2002) (Smith and Damphousse 2007) “Dataset that includes information on nearly 500 terrorists from about 60 terrorist groups indicted for more than 6,700 Federal criminal counts.” RAND Database of Worldwide Terrorism Incidents (RAND Corporation 2018) “Compilation of data from 1968 through 2009 with over 40,000 incidents of terrorism coded and detailed.” Table B-17. Source of data on terrorism.

B-44 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis Data Source Comments Center for Strategic & International Studies (CSIS) Significant cyber incidents (CSIS 2021) “This timeline records significant cyber incidents since 2006. We focus on cyber- attacks on government agencies, defense and high-tech companies, or economic crimes with losses of more than a million dollars.” Cyber Operations Tracker (Council on Foreign Relations 2021) “The Digital and Cyberspace Policy program’s cyber operations tracker is a database of the publicly known state- sponsored incidents since 2005.” VIZSEC (VizSec Organization n.d.) VizSec is an independent organization devoted to researching data mining and visualization for computer security. VizSec’s data page lists many links to data sets. National Vulnerability Database (NVD) [National Institute of Standards and Technology (NIST) n.d.] “The NVD is the U.S. government repository of standards-based vulnerability management data represented using the Security Content Automation Protocol (SCAP). This data enables the automation of vulnerability management, security measurement, and compliance. The NVD includes databases of security checklist references, security- related software flaws, misconfigurations, product names, and impact metrics.” Table B-18. Data sources for cyber security threat analysis. Figure B-23. Runout modeling of a potential rock avalanche using Flow-R software.

Literature Review B-45 Modeling the extent of hydrological events, or floodplain modeling, requires hydrological and hydraulic modeling.” FEMA (2017) defines a Hydrologic and Hydraulic (H&H) study as “the study of the movement of water, including the volume and rate of flow as it moves through a watershed, basin, channel, or man-made structure.” Hydrologic modeling determines how much water from a precipitation event will become runoff and, thus, available to flood an area, while hydraulic modeling shows where the water travels. GIS software is used to develop digi- tal flood maps, incorporating such inputs as terrain models, elevation data, and imagery. U.S. Army Corps of Engineers’ (USACE’s) HEC-RAS software (USACE n.d.a) is widely used (see Figure B-25). (Source: USGS 1971.) Figure B-24. USGS shake map for San Fernando earthquake, February 9, 1971.

B-46 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis There are deterministic and probabilistic approaches for accomplishing both to describe intensity and frequency, i.e., frequency-magnitude relationships. Deterministic models are entirely determined by their parameter values and initial conditions. A deterministic model produces a single output and is thus suitable for modeling a single event. In contrast, a proba- bilistic model outputs a distribution of multiple values after running a simulation involving hundreds or even thousands of trials. Probabilistic modeling includes Monte Carlo simulation and graphical models, such as Bayesian Belief Networks and Markov Networks. Monte Carlo simulations, Bayesian Networks, and Markov chains are statistical models that incorporate probability distributions to capture the uncertainty associated with the parameters used (Bratvold and Begg 2010). Deterministic approaches include frequentism, power laws, and extreme value probability distributions. The frequentist approach calculates the number of physical events that occur per interval of time. Frequency is usually expressed as an exceedance probability, which is defined as the probability that an event with a certain magnitude will occur within a given year. For example, if a magnitude 6 earthquake occurs once every 500 years, the frequency is 1/500 or (Source: Wikimedia Commons.) Figure B-25. USACE HEC-RAS H&H modeling software.

Literature Review B-47 0.02%. If the return period of an event is known, such as a “100-year flood,” then the frequency is simply one over the return period or 1%. With an inventory of historical data that includes the magnitude of each event, it is possible to use an extreme value distribution, such as Gumbel or Poisson, to calculate the frequency for each event magnitude. The following is an example of how to calculate flood frequency. Frequency-magnitude analysis for flood events involves relating the magnitude of a flood to its frequency of occurrence through a probability distribution. Flood magnitude is usually mea- sured in terms of discharge. Hydrologists derive flood frequency from annual peak streamflow values. The USGS’s National Water Information System web interface provides access to peak streamflow data (USGS n.d.b). Probabilistic methods to derive estimates of frequencies have also been applied to rockfall (Corona et al. 2017), debris flows (Malet and Remaitre 2015), and snow avalanches (Perona et al. 2009). Interest in modeling interactive and cascading threats has grown in recent years (Pescaroli and Alexander 2018; Girgin et al. 2019). Some risk assessment frameworks and methodologies emphasize the importance of including threat interactions in risk models, especially in cases where such interactions may exacerbate the overall risk. Generally, the study of interacting threats involves identifying hazards that trigger other hazards. Cascading risks can involve interactive threats and overlapping and compounding socioeconomic consequences. For example, the disaster that struck Japan on March 11, 2011, was an earthquake that triggered a tsunami that resulted in a nuclear meltdown at the Fukushima power plant, causing power outages, contamination, and evacuations. Some studies have identified four types of threat interactions (Gill and Malamud 2014): • Interactions where a hazard is triggered (e.g., an earthquake triggers a landslide) • Interactions where the hazard probability is increased (e.g., wildfire increases the probability of a debris flow) • Interactions where the probability of a hazard is decreased (e.g., a heavy rainstorm decreases the probability of a wildfire) • Events involving the spatial and temporal coincidence of natural hazards (e.g., a volcano erupts, followed by a typhoon, causes lahars on the volcano’s slopes) Examples of these interactions include (Gill and Malamud 2014): • Japan (1972): landslide and tsunami • Alaska (1964): earthquake and tsunami • Philippines (1991): volcano eruption and earthquake/sulfur expulsion • Guatemala (2010): tropical rainstorms and mass movements Table B-19 presents a matrix showing the interaction of avalanches, debris flows, rockfalls, landslides, floods, heavy rainfall, and earthquakes (Kappes et al. 2010). (Adapted from Kappes et al. 2010.) Table B-19. Interactive hazards.

B-48 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis 5.3 Estimating Vulnerability of Assets from Relevant Threats or Hazards Another important step for performing risk and resilience analysis is estimating vulnerabil- ity. This section discusses definitions of vulnerability and the different factors and tools used to assess vulnerability. Definitions for vulnerability vary depending on the sector that is used. In general, there are many definitions of vulnerability, but several common themes emerge: a degree of loss, an adverse effect, a susceptibility to damage, a weakness, an inability to cope, or a measure of uncertainty that consequences will occur if a threat occurs (see Table B-20). Definitions of vulnerability within the transportation sector consider system performance. In her seminal paper, Berdica (2002) defined vulnerability as “a susceptibility to incidents that can result in considerable reductions in road network serviceability.” Similarly, Mattsson and Jenelius (2015) described transport system vulnerability as “society’s risk of transport system disruptions and deg- radations.” Gao et al. (2019) asserted that vulnerability stems from the factors that can contribute to the reduction of a network’s maximum service level and that topology determines performance. In addition to the multiple definitions of vulnerability, there are also multiple methodologies that estimate the vulnerabilities of the transportation network or assets. Vulnerability assessments may Definition Sector Source “The conditions determined by physical, social, UNISDR (2004) economic and environmental factors or processes, which increase the susceptibility of a community to the impact of hazards.” “Vulnerability is the degree to which a system is IPCC (2007) susceptible to, or unable to cope with, adverse effects of climate change, including climate variability and extremes.” “Vulnerability has been defined as the degree to which a system, or part of it, may react adversely during the occurrence of a hazardous event. This concept of vulnerability implies a measure of risk associated with the physical, social and economic aspects and implications resulting from the system’s ability to cope with the resulting event.” Economics Proag (2014) “Vulnerability analysis consists of estimating the conditional likelihood a threat will have on consequences …, given that the threat occurs.” Engineering ASME (2009) “Vulnerability is defined as ‘‘the degree to which the system is susceptible to and is unable to cope with adverse effects of change.” Water Adger (2006) Anandhi and Kannan (2018) “The vulnerability of each target depends on its capacity to resist various hazard values generated by surrounding sources. Vulnerability represents the potential weakness of whole targets to the hazard generated from each source.” Construction Abunemeh et al. (2017) “Vulnerability is a weakness in system security procedures, design, implementation, internal controls, etc., that could be accidentally triggered or intentionally exploited and result in a violation of the system’s security policy.” IT Computer Security Resource Center (CSRC) Information Technology Laboratory (2020) Table B-20. Definitions of vulnerability.

Literature Review B-49 be based on stakeholder input, indicator-based desk review approaches, or engineering-based analysis. The use of each type of assessment will vary depending on the individual agency’s needs and capabilities. FHWA presents an overview of these types of vulnerability assessments in their VAAF report (Filosa et al. 2017). Based on this methodology, a spreadsheet tool named the Vulnerability Assess- ment Tool (VAST) was developed to help users estimate vulnerabilities using a scoring process. Transportation agencies widely use FHWA VAAF to assess their system’s vulnerability to climate change and extreme weather events. Figure B-26 shows the different steps of this framework. (Source: Filosa et al. 2017.) Figure B-26. FHWA Vulnerability Assessment and Adaptation Framework (VAAF).

B-50 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis The framework defines vulnerability as a function of exposure, sensitivity, and adaptive capacity (Filosa et al. 2017). To determine exposure of an asset or system to a particular threat, the use of hazard maps should be implemented. Overlaying of threat maps on the asset/system should help identify if the asset or system might be located or exposed to such a threat [e.g., a FEMA Digital Flood Insurance Rate Map (DFIRM), landslide inventory, or earthquake shake map]. Sensitivity is assessed by weighing the severity of a threat against the design characteristics and condition state of the exposed asset/system. Examples of adaptive capacity include detour length, route redundancy, the number of snowplows on hand, and the response time of maintenance crews. The FHWA sponsored 19 resilience pilot projects, between 2013 and 2015, for state DOTs and MPOs to assess the vulnerability of their transportation systems. Many of these pilot projects implemented FHWA VAAF to perform their vulnerability assessments. Some of these projects include the following: • ADOT conducted a vulnerability assessment of the state highway system to extreme weather, including higher temperatures, drought, and intense storms (Anderson 2019). • Connecticut DOT assessed their system vulnerability to flooding, including bridges and culverts. • Maine DOT conducted vulnerability assessments of sea-level rise and storm surge in six coastal towns to develop depth-damage functions and adaptation design options for indi- vidual sites. • Capital Area Metropolitan Planning Organization (CAMPO) and the City of Austin con- ducted vulnerability and risk assessments from flooding, drought, extreme heat, wildfire, and ice to nine critical assets. Currently, FHWA is also sponsoring another 11 pilot projects to estimate resilience and dura- bility to extreme weather, including the use of available tools for vulnerability and risk assess- ment (Filosa et al. 2017). In addition to vulnerabilities generated based on scoring a series of parameters, vulnerabilities are also measured by degree of loss to an exposed element-at-risk and are expressed on a scale from 0 (no damage) to 1 (total damage). Argyroudis et al. (2018) stated that vulnerability of transpor- tation infrastructure can be evaluated in terms of repair costs, life-safety impacts, or degrada- tion of performance and is related to the asset’s susceptibility to damage for a given measure of hazard intensity (e.g., force, seismic loading, ground deformation, etc.). Note that vulnerability correlates the characteristics of the element-at-risk to the hazard intensity. This relationship is analogous to the concept of sensitivity found in VAAF and the applied sciences and is expressed in terms of a damage or fragility curve. Damage states are usually based on the time it takes to restore transportation assets to their original capacity and, in some cases, on the asset’s replacement or repair cost. Example damage states for roads and railways subjected to vertical ground displacement are presented in Table B-21 (Argyroudis et al. 2018). Another example of damage includes fragility curves. Figure B-27 illustrates an example damage curve for roadways subjected to inundation. The damage factor is measured against flood depth in meters. This type of fragility or damage curve is more advanced in specific fields than in others. The fields of seismology and flood science are some of the most advanced fields, while the fields of geotechnical hazards (i.e., landslide, debris flow, subsidence, etc.) are less developed. However, some attempts have been made to quantify roadway vulnerability to landslides. For example,

Literature Review B-51 Moderate 0.05 0.10 0.08 Closed during repair works Extensive/Complete 0.10 0.30 0.20 Closed during reconstruction works Permanent Vertical Ground Displacement (m) Serviceability Typology Damage State Min Max Mean Highways Minor 0.02 0.08 0.05 Open, reduced speeds, or partially closed during repair Moderate 0.08 0.22 0.15 Closed or partially closed during repair works Extensive/Complete 0.33 0.58 0.40 Closed during repair work Railways Minor 0.01 0.05 0.03 Open, reduced speeds Table B-21. Damage states for highways and railways subjected vertical ground displacement. D am ag e Fa ct or Depth (m) Flood-Depth Damage Curve for Roads Figure B-27. Example flood-depth damage curve for roads. Winter et al. (2014), developed damage curves for debris flow and three damage states based on responses to surveys sent to experts from around the world where the degree of damage was correlated with debris flow volume (m3) (see Figure B-28). Other examples of vulnerabilities based on damage factors of probabilities are developed by engineering judgment and historical performance and damage records. Examples of these vul- nerability curves or tables can be found in the CDOT Risk and Resilience Analysis Procedure

B-52 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis (AEM Corporation 2020). These vulnerability tables were developed by subject matter experts of each field, including hydraulic, rockfall, and so forth, through a series of workshops to validate the models and variables used. Table B-22 presents one of the vulnerability tables developed for CDOT’s Risk and Resilience Procedure, where vulnerabilities range from 0 to 1. As shown in this section, vulnerability assessments may be qualitative or quantitative; how- ever, most vulnerability assessments are based on exposure to threats and how well the asset will perform under such exposure. 5.4 Types of Consequences from Identified Threats Along with threats and vulnerability assessment, consequence estimation from applicable threats is a key factor in risk and resilience analysis. Different consequences can be estimated and included in risk and resilience analysis. RAMCAP Plus defines consequences as “The outcome of an event occurrence, including immediate, short, and long-term, direct and indirect losses and effects. Loss may include human (Source: Winter et al. 2014.) Figure B-28. Damage curve for debris flow and roadway. (Source: CDOT 2020.) Table B-22. CDOT’s Risk and Resilience Procedure for roadway flood vulnerabilities.

Literature Review B-53 fatalities and injuries, monetary and economic damages, and environmental impact, which can generally be estimated in quantitative terms” (ASME 2009). Additionally, consequences can be subdivided into owner, user, operational, and safety. This section addresses the consequences component of risk analysis. Asset loss is a first-order, direct consequence of an adverse event. Sometimes referred to as the owner’s financial loss, asset loss can be expressed as a percentage of repair or replacement costs and cleanup costs when applicable (ASME 2009). The degree of loss can be quantified through the application of a damage function (fragility curve, vulnerability curve, susceptibility curve, etc.) derived empirically (e.g., percent damage suffered by a building versus flood elevation) or synthetically [e.g., through expert opinion (Kreibich and Bubeck 2013; Meyer et al. 2012)]. The percentage of damage, estimated by the damage function, can be multiplied by the replacement cost or repair cost to quantify the asset loss. Alternatively, total estimates of loss can be correlated with some measure of hazard intensity (flood depth, peak ground acceleration, cubic meters of debris, etc.). Loss of functionality is also a direct consequence. Sometimes referred to as user consequences, loss of functionality impacts users due to the additional costs incurred by travel delays (lost wages and other vehicle operating costs), additional travel distance, or drive time. The FHWA HYRISK procedure for evaluating the economic losses due to bridge pier scour includes a calculation for user consequences based on traffic volume, number of days of delay, average wages, and other factors (Stein and Sedmera 2006). The CDOT Risk and Resilience Procedure’s manual incorpo- rates a modified version of the equation from HYRISK to calculate costs incurred from partial closure (work zone user costs) (CDOT 2020). Other approaches to evaluating loss of functionality include travel demand modeling, res- toration curves, and multi-criteria index models. Travel demand models, macroscopic traffic assignment models, and traffic simulation models have been used to estimate system-level con- sequences from asset loss or disruptions (due to temporary loss of function). Disruptions have been modeled by disabling links in these networks and disallowing assigning trips to the dis- rupted links or constraining their capacity. The use of a combined travel demand model incorporating trip generation, destination, mode, and route choices was proposed to assess the long-term equilibrium effects of the closure of one or more links (Chen et al. n.d.). The consequences are calculated as the decrease of a utility-based accessibility measure. Network performance related to disruptions has been researched over the years to consider concepts such as vulnerability, reliability, robustness, accessibility, and resilience. This led to the development of associated measures and indices, including the Link Performance Index for Resilience (LPIR), which evaluates the level of resilience of individual highway sections in relation to a wider network (Calvert and Snelder n.d.). It is used to detect sections with a low resiliency ranking and to analyze the underlying characteristics that make them so. A differ- ent index, known as the Network Robustness Index, was used to identify important links in a highway network. This index for a network link is defined as the increase in user equilibrium travel time that is incurred when the link is closed, which was subsequently modified to allow a less-than-total disruption and allow for comparisons between different transport networks to calculate system-level robustness (Sullivan et al. n.d.). In addition to asset loss, damage curves can be applied to assessing the loss of functionality from disruptions. A relationship between depth of standing water and vehicle speed was developed by fitting a curve to video analysis supplemented by empirical data and literature (Pregnolato n.d.). This work enabled incorporating the water depth and vehicle speed function into existing transportation models to develop better estimates of flood-induced delays. The

B-54 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis sensitivity of assets and their vulnerability to withstand a range of disruptions has been modeled using fragility and restoration models, especially for highway bridges. David Turner developed fragility curves for eight existing bridges in Colorado, considering a single damage state corre- sponding to structural failure of the bridge superstructure due to riverine ood-induced hydro- dynamic li forces (Turner 2015). A comprehensive review of fragility and restoration models was performed for typical bridge classes for application in multi-hazard risk and resilience analyses of transportation networks in the United States (Chen et al. n.d.). Similarly, Mattsson and Jenelius (2015) compiled an extensive review of the state-of-the art for modeling the conse- quences of disruption to the transportation system (Mattsson and Jenelius 2015). Husdal (n.d.) proposed a project evaluation process using a multi-criteria analysis for non- monetary and monetary impacts. e process employs a weighted multi-criteria decision approach involving the assessment of link closures or disruptions that are assessed by the severity of the impact, allowing for the assessment of individual eects or impacts. e evaluation criteria and the incremental values shown in Table B-23 are illustrative. An example of a soware application that can model both asset loss and loss of functionality is FEMA’s HAZUS-MH add-in to ArcGIS Desktop. For example, the HAZUS-MH ood model (Multi-hazard Loss Estimation Methodology n.d.) estimates damage to transportation lifeline systems based on the vulnerabilities of the various components to inundation, scour, erosion, debris impact, and hydraulic loading. e transportation lifeline life-cycle components selected for fragility modeling are primarily bridges. Impacts on system functionality, component costs, and recovery time from damage are also considered. A commercially available tool for modeling the economic consequences of disruptions to the transportation system is REMI’s TranSight which includes travel demand modeling. In the FHWA adaptation pilots conducted by Hillsborough MPO and Resilience Tampa Bay (RTBT), the Tampa Bay Regional Planning Council (TBRPC) analyzed the economic impacts of trans- portation system disruptions from six representative projects and two extreme weather sce- narios for internal ooding and hurricane events using REMI TranSight (Version 4.0). Using outputs generated from the Tampa Bay Regional Planning Model (TBRPM) for 2045, TBRPC (Source: Husdal n.d.) Table B-23. Disruption evaluation criteria.

Literature Review B-55 modeled the potential impacts of each event disrupting selected transportation links for 1 week. Results were reported using gross regional product (GRP) and personal income (or wages) as indicators. Modeling safety (societal) consequences (i.e., losses due to fatalities or injuries) involves knowing the statistical value of human life and the probability of exposure. Risk curves for fatalities are called FN curves. FN curves display the probability of having N or more fatalities per year as a function of N. FN curves can be based on historical data, probabilistic modeling, or expert opinion. The U.S. DOT used Census of Fatal Occupational Injuries (CFOI) data and a synthesis of mul- tiple studies to estimate the statistical value of human life. In 2015, the estimate was $9.4 million (Moran and Monje 2016). The estimate is a fixed value and is not adjusted for age, gender, eco- nomic status, or other demographic characteristics (Bosworth et al. 2017). Calculating the safety risk to a drive involves estimating the likelihood of exposure. Wylie expressed exposure to risk as a function of the road cut length, traffic count, and vehicle speed (see Equation 4) (Wylie 2015). ( )= × 86,400 (Eq. 4)Exposure risk LN V where: L = Length of exposed road cut (m). N = Traffic count (AADT). V = Vehicle speed (m/s). 6 Tools for Risk and Resilience Assessment This section documents and provides an overview of publicly available risk and resilience tools, including spreadsheet models, stand-alone desktop applications, and web-based tools. 6.1 Spreadsheet Models Risk Registers: Risk registers are an extremely common way for state DOTs to assess, usu- ally qualitatively, the consequence and likelihood of an event happening. States incorporate risk registers into their TAMPs to identify priority items and correlate appropriate strategies and benefit and cost estimates. While some have specific risk management chapters, others create their risk register as an overview of critical assets. States like West Virginia and Florida conduct workshops to compile their initial risk registers (FDOT Transportation Asset Management Plan 2019; WV DOT Transportation Asset Management Plan 2019). Often, an Excel spreadsheet is used, recording the likelihood and severity of consequences for each risk. Typically, a 5 × 5 matrix is employed, intersecting likelihood with severity to calculate a risk ranking. FHWA’s Center for Innovative Finance Support provides an example of a qualitative risk matrix (see Figure B-29) (FHWA 2015). O’Har et al. (2016) detail developing a risk register. VAST: The U.S. DOT’s tool is an Excel spreadsheet implementation of the FHWA VAAF (Bhat et al. 2015). The tool is designed to help planners at state DOTs and MPOs assess the vulner- ability of their transportation assets to climate stressors. It is an indicator-based tool that assigns an index value to three factors: • Exposure • Sensitivity • Adaptive capacity

B-56 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis Index values for the three factors are summed to calculate a composite risk score. PennDOT developed its spreadsheet tool to implement VAAF as part of its Extreme Weather Vulnerability Study (PennDOT 2017). Relative index values representing factors such as AADT and functional class are manually entered for each roadway segment. The inputted values are multiplied by the weights, shown in Figure B-30. HYRISK: Sponsored by the FHWA, HYRISK is an Excel-based, scour risk assessment tool for bridges. The tool requires inputs from the National Bridge Inventory (NBI) to calculate the probability of failure due to scour (see Table B-24) (Pearson et al. 2000). HYRISK uses the fol- lowing NBI Items: • Item 26: Functional class • Item 43: Structure type • Item 60: Substructure condition rating • Item 61: Channel and channel protection condition rating • Item 71: Waterway adequacy • Item 113: Scour-critical bridges CDOT’s Risk and Resilience Excel Spreadsheet Tool: CDOT developed a Risk and Resil- ience Analysis Procedure to calculate annual risk to highway infrastructure from the following threats: flood, rockfall, and scour. Based on this procedure, CDOT developed an Excel Spread- sheet tool (CDOT 2020). The CDOT Risk and Resilience tool is a deterministic model and calculates annual owner and user risk for roads, bridges, and culverts. 6.2 Software Tools Some software tools are also available to estimate risk and resilience to certain assets from certain threats. However, these tend to be more data-intensive and require higher knowledge and capabilities. Some of these tools include: HAZUS-MH: This tool, considered the state-of-the art risk tool for natural hazard risk assess- ment, is a GIS-based tool provided freely to the public by the Federal Emergency Manage- ment Administration (FEMA). However, the tool is an add-in to ESRI’s ArcGIS Desktop application, requiring a licensing fee. The latest version, 4.0, was released in 2017. The HAZUS earthquake model includes fragility curves for roads and bridges (FEMA 2020a). The HAZUS flood model can generate depth grids that can estimate flood extent and the likelihood of overtopping (FEMA 2018). The flood model also includes a scour risk model (Source: FHWA 2015.) Figure B-29. Qualitative risk matrix.

Literature Review B-57 (Source: PennDOT 2017.) Figure B-30. PennDOT extreme weather vulnerability risk assessment tool. Table B-24. Bridge probability from scour using HYRISK model. (Source: Pearson et al. 2000.)

B-58 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis for bridges but does not include models to address damage to road surfaces. Quantitative outputs for the earthquake model include estimates of direct losses, calculated as a fraction of the replacement value. For the flood model, estimated losses are computed for bridges but not roads. HAZUS-MH is also a resilience assessment tool in that it includes restora- tion curves for bridges in the earthquake (FEMA 2020a) and flood models (FEMA 2018). The earthquake restoration curve is based on expert opinion and conforms to a normal cumulative distribution function. The flood restoration curve is based on damage state versus time to rebuild. REDARS 2: Developed by MCEER under the sponsorship of FHWA, REDARS 2 stress tests bridges, tunnels, and retaining structures under seismic loading. The software includes the capability to simulate traffic volume, pre-, and post-event (Werner et al. 2006). 6.3 Web-Based Application Tools Besides spreadsheets and software, there are also web-based tools or models to calculate risk. Examples of these tools are: Vermont Transportation Resilience Planning Tool (TRPT): The Vermont TRPT is a web map that enables the user to visualize relative flood risk to embankments, culverts, and bridges. The potential risk is estimated based on criticality and asset vulnerability and proposes mitigation alternatives to reduce risk. The level of risk is indicated by color coding roadway segments: low (green), medium (orange), and high (red). Low levels of vulnerability may mean slight damage due to inundation or minor erosion, whereas high vulnerability may mean complete failure of the asset due to severe erosion or deposition (VTRANS n.d.). Figure B-31 shows a representation of assets and their different risk levels. Boston Harbor Flood Risk Model (BH-FRM): This is a dynamic flood model developed by Massachusetts DOT (MassDOT) under FHWA’s Climate Resilience pilots (Massachusetts DOT n.d.). This model was developed to determine inundation risk and flooding pathways. This tool simulates the effects of tides, storm surge, wind, waves, river discharge, sea-level rise, and climate scenarios. ESRI’s Resilience Dashboard: This is a preconfigured web-based dashboard that runs in ArcGIS Online (ESRI’s cloud). Designed for emergency management, the dashboard includes drop-down windows to enable users to filter the features displayed by asset class and a limited number of attributes relevant to resilience and criticality, such as “importance to the community – high” or “dependent upon flood pumps – yes.” The tool displays an overall resilience ranking, aggregated by county (ESRI n.d.). Many transportation agencies are currently developing their models and tools to estimate risk and resilience to different assets and threats. Some agencies are more advanced in this process than others. In addition, some of these initiatives are developed based on sponsor- ships from FHWA or TRB, while other agencies invest in their resilience programs to build these tools. The following tools are web tools developed and hosted by the USACE. Sea-Level Change Calculator: The USACE Sea-Level Change Calculator generates sea-level change projections for a user-selected NOAA sea-level gauge (USACE 2017a). The user can print out the curve of the change in sea level over time, either as a graph or table. Sea-Level Tracker: The USACE Sea-Level Tracker tool “does not predict future water levels. Rather, the tool offers smoothed analysis of historical sea-level behavior and the measured trends at a user-selected gauge” (USACE 2017b). The tool includes a map window, data visu- alization window (trend line), and data table view. The user first selects a location and then

(Source: VTRANS n.d.) Figure B-31. Screenshot of Vermont TRPT.

B-60 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis a datum: [Mean Sea Level (MSL), Mean High Water (MHW), or Mean of the Higher High- Water (MHHW) height]. e output is a graph of the trend line from 1975 to the present. Climate Hydrology assessment Tool: e USACE Climate Hydrology Assessment Tool gen- erates a trend line showing the change in peak streamow based on climate change projec- tions (USACE n.d.b). 6.4 Methods for Developing Recovery Strategies Recovery refers to the time it takes for an asset, corridor, or system to return to normal aer a disruption. e National Disaster Recovery Framework denes four phases: (1) preparedness (pre-disaster), (2) short-term (days), (3) intermediate (weeks to months), and (4) long-term (months to years) (FEMA 2020b) (see Figure B-32). e time it takes for a system to recover is one (Source: FEMA 2020b.) Figure B-32. National Disaster Recovery Framework.

Literature Review B-61 possible metric for resilience, because it measures how well planners are meeting the needs of the public to resume the use of damaged infrastructure. On this theory, academic studies have devel- oped measure of resilience (MOR) models that incorporate recovery time. Two studies develop these models on highways using freight metrics. The first modeled the ability of intermodal freight networks to withstand and recover from disruptors (Chen and Miller-Hooks 2011), and the second studied the reduction of service and recovery levels to pre-event conditions along the I-90/I-94 corridor in Wisconsin during two weather events (Adams et al. 2012). Although these studies provide a theoretical framework for understanding recovery, a gap remains between this theory and the practical strategies planners can use to quicken recovery after a hazard event. One source for these strategies is a report from AASHTO released in May 2018 on Resiliency Case Studies: State DOT Lessons Learned (WSP n.d.). This report reviewed eight extreme weather events, including tropical storms, hurricanes, flooding events, rock falls, ice storms, tornadoes, and landslides, and cataloged how state DOTs responded to these events. In particular, it contained lessons on the emergency response relevant to recovery strategies in the aftermath of a hazard event. This report suggested the following steps that DOTs could take: • Enter into contracts and memorandums of agreements (MOAs) before any emergency to assist with the response and expand department capacity. This could include areas such as debris removal and flooding response. • Use social media during emergencies to disseminate information directly to the public, in addition to traditional mediums such as TV, radio, and electronic signage. • Prior to an emergency, have a financial accounting system to maximize access to and benefits from relief funds. • Work with emergency management departments to have a centralized emergency manage- ment center to streamline the response. • Set up partnerships with GPS and mapping applications to disseminate information about road closures and detours. • Prioritize public safety in all parts of an emergency response. Additionally, NCHRP Research Report 931: A Guide to Emergency Management at State Trans- portation Agencies places transportation agencies within the larger context of emergency response (Frazier et al. 2020). Using a framework known as the Incident Command System, this guide explains the roles and responsibilities likely to face transportation agencies during an emergency. It provides details and examples of policies and plans that the agencies should have to help them prepare for coordinated emergency response. Finally, NCHRP Report 753: A Pre-Event Recovery Planning Guide for Transportation helps agencies with their pre-event planning process that can assist in developing recovery strategies (Bye et al. 2013). This guide emphasizes use of informal and formal networks, pre-existing plans, and preventative infrastructure improvements in recovery strategies. It also summarizes how pre-event planning can be a part of other planning processes, such as asset management and risk management plans. 6.5 Methods for Risk Management Risk management is the process of identifying, measuring, managing and mitigating risks that might threaten goals and initiatives that an organization pursues. It is a natural complement to performance management and asset management. Where agencies use those two tools to set goals and initiatives, agencies use risk management to catalog and mitigate factors that may prevent those goals and initiatives from succeeding.

B-62 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis Risks are uncertain events that, if they occur, can affect at least one goal or objective of a project, program, or enterprise (Risk Management Guidance n.d.). Risks can either be negative (and are called threats) or positive (called opportunities). Strategies to manage threats and hazards, such as avoiding, transferring, or mitigating the risk, reduce the likelihood that it will occur and the impact it will have if it occurs. Strategies to manage opportunities, such as exploiting, sharing, or enhancing the risk, take the opposite approach to threats and seek to make them more likely to occur and/or increase the impact if it occurs. At a certain point, however, agencies must accept the risks, whether they are threats or opportunities, and direct planners to track and monitor those risks. The final report from NCHRP Project 08-93 is a risk management guide for state DOTs (Proctor et al. 2016). It gives DOTs detailed guidance for establishing an enterprise risk management policy tailored to their environment. Some benefits provided by risk management that this guide emphasizes include financial strategies because agencies benefit from clear processes used to allocate resources, and communication benefits, because agencies can use risk as a framework to explain goals and implement strategies. This guide contains many risk management strategies and applies them to different situations that a state DOT may face. A selection of some of the significant strategies covered in the guide include: • Put an enterprise risk management process in place that (1) incorporates levels of risk (from strategic risks across the enterprise to risks associated with specific department activities); (2) provides staff with the tools and training to implement the process; and (3) integrates risk management into agency processes at regular intervals. • Use various techniques to identify risks, such as brainstorming workshops with staff, structured interviews, the development of checklists, and step-by-step process reviews. • As risks are identified, examine the environment surrounding that risk, documenting both the external factors that may influence an objective and the internal environment with different components involved in responding to that risk. • Organize risks into categories such as health and safety, occupational, economic, political, regulatory, information, natural environment, fraud or malfeasance, and litigation risks. • Once categorized, determine the cause of risks, which can be done through workshops or simulations, focusing on determining the root cause. Other sections of this literature review cover additional strategies covered in the guide, such as risk registers and quantitative methods for estimating risk consequences and likelihood. 6.6 Methods to Improve System Resilience Resilience is defined by the FHWA as “the ability to anticipate, prepare for, and adapt to changing conditions and withstand, respond to, and recover rapidly from disruptions (FHWA 2014).” The FHWA Office of Planning, Environment, and Realty maintains a website on resil- ience (Resilience n.d.). It is in line with the FHWA’s work to increase the health and longevity of the nation’s highways by assessing vulnerabilities, considering resilience in the transportation planning process, incorporating resilience in asset management plans and long-range transpor- tation plans, addressing resilience in project development and design, and optimizing opera- tions and maintenance practices. This website maintains guidance, framework and pilots, and research projects to support this work. One of the resources on this website is a white paper summarizing how state DOTs and MPOs incorporate resilience into the transportation planning process (Dix et al. 2018). This document summarizes how these organizations define resilience, which agencies incorporate resilience into their planning process, and why and how they do so. It is this last question, the “how,”

Literature Review B-63 where there is the most information. The paper provides a variety of perspectives on the goals of resilience planning, the assessment of problems or needs, performance measures and targets set out in the plans, and strategies available (including policy-based, flooding-related, opera- tional, and partnership-based strategies). However, the paper notes that substantive information regarding the implementation of strategies is not yet available. Resilience strategies specific to highway systems generally revolve around two categories: (1) design guidelines; and (2) flooding or stormwater management. Design guidelines can imple- ment policy-based strategies, such as how climate change may impact vertical clearance for bridges or the monitoring of pavement conditions. Flooding and stormwater management strategies reflect concerns about how the highway system will respond to hazards, such as significant storms, or shifting environmental conditions, such as increases in river flow or rising sea levels. The AASHTO report on resiliency case studies notes the importance of updating hydraulics manuals and strategic plans (WSP n.d.). Overall, incorporating climate risk and variability into asset management practices and prioritization of resources to maintain and reinforce the most vulnerable and critical assets can improve trans- portation system resilience. FEMA maintains an Office of Resilience Integration and Coordination to ensure that agency activities across Resilience are unified and coordinated with the FEMA regional offices, other federal agencies, and partnering industry associations. The Resilience Office also provides train- ing through the National Preparedness Directorate for the nation’s first responders and helps communities prepare for, respond to, and recover from disasters (Resilience n.d.). RAND Corporation (n.d.) suggests planning for disruptions to improve system resilience, including detour management plans and continuity of operations plans. The report also sug- gests developing tools to assess the dynamic performance of integrated passenger and freight transportation systems by incorporating resilience into travel demand models. Another critical consideration includes incorporating adaptive capacity to determine system capacity and redun- dancies while preparing plans to reallocate resources as required to address potential disruptions. Consistent and reliable funding is also a key challenge when improving system resilience. Programs that fund highway expenditures, such as the National Highway Performance Program and the Surface Transportation Block Grant Program, make expenditures that fund improved system resilience eligible for funding. The Emergency Relief Program funds can also be used on repairs that will enhance the resilience of federal-aid highways (FHWA n.d.b). To be approved, state DOTs must demonstrate that those repairs are economically justified by weighing the cost of improvements against the risk of recurring damage and the cost of future repair. 7 Tracking Mechanisms for Risk and Resilience Metrics Various types of tracking mechanisms have been developed to evaluate and process risk and resilience metrics. States use these mechanisms to prepare risk management efforts and incor- porate resilience into their systems. In addition to traditional tracking methods such as written reports and spreadsheets, transportation agencies rely more on GIS to produce vivid story maps and dashboards. 7.1 Recurring Reports Annual reports help show the yearly impact of various assets and allow states to track progress and implement better mitigation efforts moving forward. Minnesota’s “MinnesotaGo” program is a program aligning the MnDOT system with the public regarding quality of life, economy,

B-64 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis and the natural environment (MnDOT 2022). They developed a “Trend Library” that showcases metrics and annual updates, allowing the continual impact over the years to be easily understood. It should be noted that these various tracking mechanisms are shaped for the agency, reflect- ing their characteristics, priorities, and data, but they can be transferable and utilized by other agencies to improve their own risk and resilience metrics. Snapshot observations of these metrics that are memorialized, such as in an annual report, are also used in public communication initiatives, as content for brochures, and white papers to inform the public more strategically on agency consideration of risks and resilience measures. An example of such a communication is the 2019 “Why Should We Care About Roads?” brochure created by VTRANS (2019). This format can effectively communicate core values and highlight goals or progress. 7.2 Story Maps ArcGIS StoryMaps is an ESRI proprietary web application that enables users to integrate digital maps with narrative content and other multi-media. ESRI’s Story Map Series Gallery includes examples of story maps by state and local transportation agencies (ESRI 2021). 7.3 Dashboards In addition to map views, GIS web applications can also include widgets that can dynamically display quantitative measures such as traffic counts or accidents. The Hampton Roads Resilience Project dashboard (see Figure B-33) consists of a map view of point locations for projects, sym- bolized by color by project types, such as drainage improvements or stormwater management. To the left, there are widgets to display quantitative data in terms of cost in millions of dollars by project type and status. Another example of a dashboard is the Southeast Michigan Council of Government’s (SEMCOG) Flooding Risk Tool Dashboard [Southeast Michigan Council of Governments (SEMCOG) and Michigan DOT (MDOT) 2020] (see Figure B-34). In the map view, road seg- ments are symbolized by color by risk rating: low (green), medium (gold), and high (red). The widgets to the right include a pie chart illustrating the percentage of roads rated as low, medium, or high risk; the total number of road assets, bridges, culverts, and pump stations; and a list of the top five road segments at risk with associated vulnerability and criticality scores. 8 Barriers and Constraints for Implementation of Quantitative Risk and Resilience Assessment Methods Barriers to implementing risk and resilience methods include the lack of robust evaluation and validation processes, availability of supporting data and tools, and limited technical capacity and knowledge base to implement quantitative risk and resilience assessments. Among the barriers to incorporating resilience assessments in transportation planning are the lack of evaluation methods and criteria for monitoring existing plans that use resilience strategies. The review of how DOTs and MPOs use resilience assessments in their planning processes revealed that only a limited number of DOTs and MPOs include resilience-related performance measures (Dix et al. 2018). Such perfor- mance measures have not been tracked for long enough to deduce any lessons on monitoring and reporting. Among those agencies that have implemented resilience assessments, their monitoring metrics relate to the frequency of these assessments being updated. Therefore, a review of how states are monitoring the implementation of resilience assessments remains a significant gap.

Literature Review B-65 Conducting quantitative risk and resilience assessments requires a substantial amount of data. A possible barrier to implementation is the management and collection of these data types in a centralized and systematic database. For example, the FHWA’s report on Vulnerability Assess- ment and Adaptation Framework notes that complete data may not be available for all assets (Filosa et al. 2017). Though a considerable amount of data is available along with major systems such as the NHS, more granular data, such as culvert slope, as-built information, and flood his- tory, may not be available for roads outside the NHS. This can hinder vulnerability assessments for minor arterial and collector roads. A more systematic and comprehensive review of recovery strategies can help return condi- tions to normal after a hazard impacts service. There has been ample preparation for hazards in (Source: Hampton Roads Transportation Planning District Commission, 2014.) Figure B-33. Hampton Roads Resilience Project dashboard.

B-66 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis emergency management, including guidebooks and exercises to make plans in collaboration with emergency management departments. However, pre-event recovery strategies for state DOTs are needed. NCHRP Report 753: A Pre-Event Recovery Planning Guide for Transportation was published in 2013 (Bye et al. 2013). Although it does include strategies for developing networks and plans with other agencies, it does not incorporate the progress made in topics like asset man- agement, performance management, and risk assessments that have been accomplished since that time. These types of plans can be used to help improve the post-hazard recovery response. Finally, numerous risk and resilience analysis tools have been developed in recent years that vary significantly in their complexity. Though this represents an advancement in the sophistication of [Source: Southeast Michigan Council of Governments (SEMCOG) and Michigan DOT (MDOT) 2020.] Figure B-34. SEMCOG Flooding Risk Tool dashboard.

Literature Review B-67 these techniques, it can also act as a barrier to implementation. For example, in the final report of NCHRP Project 08-93, Proctor et al. (2016) explained complex risk management tools were discontinued because of their complexity and lack of transparency. The effort to develop increas- ingly sophisticated risk assessment tools needs to be balanced with ease of use for end-users and transparency about how these tools work. Quantitative risk and resilience methods have been researched over the past two decades, with only a few techniques successfully adopted into applied practice. Standardized and well- understood methods may need to be adopted using cooperative research efforts and coordina- tion among academic researchers and practitioners to enable the transition for a widespread application. References AASHTO. (2013). AASHTO Transportation Asset Management Guide—Executive Summary. Retrieved from https://www.fhwa.dot.gov/asset/pubs/hif13047.pdf. AASHTO. (2016a). Guide for Enterprise Risk Management: Quick Guide. AASHTO. (2016b). Understanding Transportation Resilience: A 2016–2018 Roadmap for Security, Emergency Management, and Infrastructure Protection in Transportation Resilience. Washington, DC. Abunemeh, M., EL Meouche, R., Hijaze, I., Mebarki, A., and Shahrour, I. (2017). Hazards, Vulnerability and Interactions at Construction Sites: Spatial Risk Mapping. Journal of Information Technology in Construction, 22(4), 63–79. Adams, T. M., Bekkem, K. R., and Toledo-Duran, E. J. (2012). Freight Resilience Measures. Journal of Transporta- tion. Retrieved from https://ascelibrary.org/doi/abs/10.1061/%28ASCE%29TE.1943-5436.0000415. Adaptation Clearing House. (2015). U.S. Department of Transportation’s Vulnerability Assessment Scoring Tool (VAST). Retrieved from https://www.adaptationclearinghouse.org/resources/u-s-department-of- transportation-s-vulnerability-assessment-scoring-tool-vast.html. Adger, W. N. (2006). Vulnerability. Global Environmental Change, 16(3), 268–281. ADOT. (2019, February). Asset Management, Extreme Weather, and Proxy Indicators Pilot Project. Retrieved from https://azdot.gov/sites/default/files/2019/07/Asset- Mgmt-Extreme-Weather-and-Proxy-Indicators- Pilot-Project.pdf. AEM Corporation. (2017, November 30). I-70 Corridor Risk & Resilience Pilot Final Report. Retrieved from https://www.codot.gov/programs/planning/documents/plans-projects-reports/reports/i70rnr_final report_ nov302017_submitted_af.pdf. AEM Corporation. (2020). Colorado Department of Transportation Risk and Resilience Procedure: A Manual for Calculating Risk to CDOT Assets from Flooding, Rockfall, and Fire Debris Flow. Retrieved from https://www. codot.gov/programs/planning/cdot-rnr-analysis-procedure-8-4-2020-v6.pdf. Alder, R., Wirth, J., Moser, A., Busch, G., Henrie, J., and Slooten, N. V. (2020). UDOT Asset Risk Management Process. Salt Lake City, UT: Utah Department of Transportation. Allen, B. (2020). NCHRP Synthesis 556: Asset Management Approaches to Identifying and Evaluating Assets Damaged Due to Emergency Events. Transportation Research Board, Washington, DC. Anandhi, A., and Kannan, N. (2018). Vulnerability Assessment of Water Resources—Translating a Theoretical Concept to an Operational Framework Using Systems Thinking Approaching a Changing Climate: A Case Study in Ogallala Aquife. Journal of Hydrology, 557, 460–474. Anderson, T. (2019). ADOT Transportation Asset Management Plan. Phoenix, AZ: Arizona Department of Transportation. Argyroudis, S., Mitoulis, S., Kaynia, A. M., and Winter, M. G. (2018). Fragility Assessment of Transportation Infrastructure Systems Subjected to Earthquakes. Geotechnical Earthquake Engineering and Soil Dynamics V: Numerical Modeling and Soil Structure Interaction, 174–183. Asam, S., Bhat, C., Dix, B., Bauer, J., and Gopalakrishna, D. (2015). Climate Change Adaptation Guide for Transportation Systems Management, Operations and Maintenance. Washington, DC: U.S. Department of Transportation, Federal Highway Administration. Retrieved from http://www.dot.state.oh.us/Divisions/ Operations/Traffic-Management/TSMO%20Docs/FHWA%20Climate%20Change%20Adaptation%20 Guide%20for%20Transportation%20Systems%20Management%20Operations%20and%20Maintenance.pdf. ASME. (2009). All-hazards Risk and Resilience: Prioritizing Critical Infrastructures Using the RAMCAP Plus Approach. New York: ASME Innovative Technologies Institute, LLC. AWWA. (2010). J100-10 RAMCAP Standard for Risk and Resilience Management for Water and Wastewater. Denver, CO: American Water Works Association.

B-68 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis Battelle. (2007). Evaluation of System’s Available Redundancy to Compensate for Loss of Transportation Assets Resulting From Natural Disaster or Terrorist Attacks. Report of the National Surface Transportation Policy and Revenue Study Commission: Transportation for Tomorrow. National Surface Transportation Policy and Revenue Study Commission. Berdica, K. (2002). An Introduction to Road Vulnerability: What Has Been Done, Is Done and Should Be Done. Transport Policy, 9, 117–127. Bhat, C., Hyman, R., Kafalenos, R., Beucler, B., Choate, A., and Rodehorst, B. (2015, September 17). FHWA Vulnerability Tools. Retrieved December 9, 2020, from http://onlinepubs.trb.org/onlinepubs/conferences/2015/ ClimateChange/67.CassandraBhat.pdf. Bosworth, R. C., Hunter, A., and Kibria, A. (2017). The Value of a Statistical Life: Economics and Politics. Logan, UT: Strata. Bratvold, R. B., and Begg, S. H. (2010). Making Good Decisions. Richardson, TX: Society of Petroleum Engineers. Bruneau, M., and Reinhorn, A. (2006). Overview of the Resilience Concept. National Conference on Earthquake Engineering. San Francisco. Bruneau, M., Chang, S. E., Eguchi, R. T., and Lee, G. C. (2003). A Framework to Quantitatively Assess and Enhance the Seismic Resilience of Communities. Earthquake Spectra, 19, 733–752. Bye, P., Yu, L., Shrivastava, S., and van Leeuwen, S. (2013). NCHRP Report 753: A Pre-Event Recovery Planning Guide for Transportation. Transportation Research Board of the National Academies, Washington, DC: Retrieved from https://nap.nationalacademies.org/catalog/22527/a-pre-event-recovery-planning-guide- for-transportation. Caltrans. (2018, April 11). Environmental Risk Register Guidance. Retrieved from https://dot.ca.gov/-/media/ dot-media/programs/environmental-analysis/documents/ser/f0005505-env-risk-mngmt-a11y.pdf. Caltrans. (2019, May). California Transportation Asset Management Plan Fiscal Years 2017/18-2026/27. Retrieved from https://dot.ca.gov/-/media/dot-media/programs/asset-management/documents/20190726-am-final californiatamp-a11y.pdf. Caltrans Division of Research, Innovation and System Information. (2015, July 22). Application of Quantita- tive Risk Management Practices to Project Development: A Survey of State Practice and Related Research. Retrieved from https://dot.ca.gov/-/media/dot-media/programs/research-innovation-system-information/ documents/preliminary-investigations/application-of-quantitative-risk-management-practices-pi-a11y.pdf. Calvert, S., and Snelder, M. (n.d.). A Methodology for Road Traffic Resilience Analysis and Review of Related Concepts. Retrieved from https://www.tandfonline.com/doi/full/10.1080/23249935.2017.1363315. Cambridge Systematics. (2011). Climate Change Vulnerability and Risk Assessment of New Jersey’s Transporta- tion Infrastructure. Retrieved from https://www.njtpa.org/NJTPA/media/Documents/Planning/Regional- Programs/Studies/Climate%20Change%20Vulnerability%20and%20risk%20Assessment/CCVR_ REPORT_FINAL_4_2_12_ENTIRE.pdf?ext=.pdf. Cambridge Systematics and ICF International. (2015). Central Texas Extreme Weather and Climate Change Vulnerability Assessment of Regional Transportation Infrastructure—Final Report. Retrieved from http:// austintexas.gov/sites/default/files/files/CAMPO_Extreme_Weather_Vulnerability_Assessment_FINAL.pdf. Cape Cod MPO. (2015). Cape Cod 2016 Regional Transportation Plan 2016–2040. Barnstable, MA: Cape Cod Metropolitan Planning Organization. CDOT. (2019). Risk-Based Asset Management Plan Version 2. Colorado Department of Transportation. CDOT. (2020). CDOT Resilience Program. Retrieved December 10, 2020, from https://www.codot.gov/programs/ planning/cdot-resilience-program. Chen, A., Yang, C., Kongsomsaksakul, S., and Lee, M. (n.d.). Network-based Accessibility Measures for Vulnerability Analysis of Degradable Transportation Networks. Retrieved from https://www.researchgate.net/publication/ 5153107_Network-based_Accessibility_Measures_for_Vulnerability_Analysis_of_Degradable_Transportation_ Networks. Chen, L., and Miller-Hooks, E. (2011, August 18). Resilience: An Indicator of Recovery Capability in Intermodal Freight. Retrieved from https://pubsonline.informs.org/doi/10.1287/trsc.1110.0376. Chen, S., Ioannis, G., Jamie, P., and Barbosa, A. (n.d.). Multiple-Hazard Fragility and Restoration Models of Highway Bridges for Regional Risk and Resilience Assessment in the United States: State-of-the-Art Review. Retrieved from https://www.researchgate.net/publication/308755769_Multiple-Hazard_Fragility_and_ Restoration_Models_of_Highway_Bridges_for_Regional_Risk_and_Resilience_Assessment_in_the_ United_States_State-of-the-Art_Review Computer Security Resource Center (CSRC) Information Technology Laboratory. (2020). Vulnerability. Retrieved January 13, 2020, from https://csrc.nist.gov/glossary/term/vulnerability. Corona, C., et al. (2017). Modeling Rockfall Frequency and Bounce Height from Three-Dimensional Simulation Process Models and Growth Disturbances in Submontane Broadleaved Trees. Geomorphology, 281, 66–77. Council on Foreign Relations. (2021). Cyber Operations Tracker. Retrieved from https://www.cfr.org/cyber- operations/.

Literature Review B-69 CSIS (Center for Strategic & International Studies). (2021). Significant Cyber Incidents. Retrieved from https:// www.csis.org/programs/strategic-technologies-program/significant-cyber-incidents. Curtis, J. A., et al. (2012). Transportation Risk Management: International Practices for Program Development and Project Delivery. FHWA. Retrieved from https://international.fhwa.dot.gov/scan/12029/12029_report.pdf. DHS. (2009). National Infrastructure Protection Plan. Washington, DC: U.S. Department of Homeland Security. DHS. (2010). Risk and Resilience: Exploring the Relationship. Arlington, VA: U.S. Department of Homeland Security. DHS. (2012). Threat and Hazard Identification and Risk Assessment Guide. Washington, DC: U.S. Department of Homeland Security. Dix, B., Zgoda, B., Vargo, A., Heitsch, S., and Gestwick, T. (2018). Integrating Resilience into the Transporta- tion Planning Process: White Paper on Literature Review Findings. Washington, DC: U.S. Department of Transportation, Federal Highway Administration. Downscaled CMIP3 and CMIP5 Climate and Hydrology Projections. (n.d.). Retrieved from https://gdo-dcp. ucllnl.org/downscaled_cmip_projections/. Duke University Libraries. (2018). International Terrorism: Attributes of Terrorist Events (ITERATE). Retrieved from https://library.duke.edu/data/sources/iterate. Engineering Safety Consultants. (n.d.). Quantitative Risk Assessment—QRA. Retrieved from https://esc.uk.net/ quantitative-risk-assessment/#:∼:text=A%20Quantitative%20Risk%20Assessment%20(QRA,company%20 assets%20and%20its%20reputation. ESRI. (2021). Story Map Series Gallery. Retrieved from https://storymaps-classic.arcgis.com/en/app-list/map- series/gallery-series/#s=30&q=highway. ESRI. (n.d.). Resilience Dashboard. Retrieved December 9, 2020, from https://solutions.arcgis.com/emergency- management/help/resilience-dashboard/. FDOT Transportation Asset Management Plan. (2019). Retrieved from https://fdotwww.blob.core.windows. net/sitefinity/docs/default-source/planning/performance/fdot-transportation-asset-management-plan_ (june-28-2019).pdf?sfvrsn=36c94a6b_2. Federal Railroad Administration. (2015). Risk Evaluation Framework and Selected Metrics for Tank Cars Carrying Hazardous Materials. Washington, DC: U.S. Department of Transportation, Federal Railroad Administration. FEMA. (2017, October). Hydrologic and Hydraulic (H&H) Study Quick Guide. Retrieved from https://www. sog.unc.edu/sites/www.sog.unc.edu/files/general_media/FEMA%20HH%20Study%20Quick%20 Guide%28Final%29.pdf. FEMA. (2018). Hazus Flood Model User Guidance. Washington, DC. FEMA. (2020a). Hazus Earthquake Model Technical Manual. Hazus 4.2 SP3. Washington, DC. FEMA. (2020b, July 23). National Disaster Recovery Framework. Retrieved from https://www.fema.gov/emergency- managers/national-preparedness/frameworks/recovery. FEMA. (2021, August 6). National Flood Hazard Layer. Retrieved from https://www.fema.gov/flood-maps/ national-flood-hazard-layer. Ferguson, W. S. (2002). Hurricane Preparedness and Recovery by a Transportation Agency. Charlottesville, VA: Virginia Transportation Research Council. FHWA. (2010). Advancing Metropolitan Planning for Operations: The Building Blocks of a Model Transporta- tion Plan Incorporating Operations. Washington, DC: U.S. Department of Transportation, Federal Highway Administration. FHWA. (2012). Risk-Based Transportation Asset Management: Achieving Policy Objectives by Managing Risks. Report 3: Risks to Asset Management Policies. Washington, DC: U.S. Department of Transportation, Federal Highway Administration. Retrieved from https://www.fhwa.dot.gov/asset/pubs/hif12054.pdf FHWA. (2013, September 12). MAP-21 Moving Ahead for Progress in the 21st Century. Retrieved from https:// www.fhwa.dot.gov/map21/factsheets/pm.cfm. FHWA. (2014, December 15). FHWA Order 5520. Retrieved from https://www.fhwa.dot.gov/legsregs/directives/ orders/5520.cfm. FHWA. (2015, March 7). P3 Project Risk Assessment: P3-VALUE Webinar. Retrieved from https://www.fhwa.dot. gov/ipd/pdfs/p3/p3-value_2.0_session_4_ra_presentation_030716.pdf. FHWA. (2017, November). Incorporating Risk Management into Transportation Asset Management Plans. Retrieved from https://www.fhwa.dot.gov/asset/pubs/incorporating_rm.pdf. FHWA. (2020a). CMIP Climate Data Processing Tool 2.1. Retrieved from https://fhwaapps.fhwa.dot.gov/cmip. FHWA. (2020b). Every Day Counts Initiative. https://www.fhwa.dot.gov/innovation/everydaycounts/. FHWA. (n.d.a). FHWA Sustainability—2018—2020/2024 Pilot Program: Resilience and Durability to Extreme Weather. Retrieved January 31, 2021, from https://www.fhwa.dot.gov/environment/sustainability/resilience/ pilots/extweatherpilot.cfm.

B-70 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis FHWA. (n.d.b) Federal-aid Programs and Special Funding. Retrieved from https://www.fhwa.dot.gov/specialfunding/ er/191011.cfm#_ftn2. FHWA. (n.d.c). Integrating Resilience into the Transportation Planning Process—White Paper on Literature Review Findings. Retrieved from https://www.fhwa.dot.gov/environment/sustainability/resilience/ongoing_and_ current_research/planning/integrating_resilience.cfm#toc511390376. FHWA, Kalla, and Shepherd. (n.d.). Integration of Resilient Infrastructure in the Emergency Relief Program. Retrieved from https://www.fhwa.dot.gov/specialfunding/er/191011.cfm. Filosa, G., Plovnick, L., Miller, R., and Pickrell, D. (2017). Vulnerability Assessment and Adaptation Framework, 3rd ed. FHWA. Retrieved from https://www.fhwa.dot.gov/environment/sustainability/resilience/adaptation_ framework/climate_adaptation.pdf. Fixing America’s Surface Transportation Act (FAST Act). (2015). Retrieved from https://www.govinfo.gov/con- tent/pkg/PLAW-114publ94/pdf/PLAW-114publ94.pdf. Flannery, A. (2019, November 13). Development and Incorporation of Quantitative Risk and Resilience Analysis Standards into Agency Decision Making. Retrieved from https://trb.secure-platform.com/a/solicitations/15/ sessiongallery/197/application/865. Flannery, A., Pena, M. A., and Manns, J. (2018). NCHRP Synthesis 527: Resilience in Transportation Planning, Engineering, Management, Policy, and Administration. Washington, DC: Transportation Research Board, Washington, DC. doi:10.17226/25166. Frazier, E. R. Sr., Y. J. Nakanishi, P. Auza, J. L. Western, P. G. Bye, and D. Matherly. (2020). NCHRP Research Report 931: A Guide to Emergency Management at State Transportation Agencies. TRB, Washington, DC. Retrieved from https://www.nap.edu/catalog/25557/a-guide-to- emergency-management-at-state- transportation-agencies. Freckleton, D., Heaslip, K., Louisell, W., and Collura, J. (2012). Evaluation of Resiliency of Transportation Networks after Disasters. Transportation Research Record: Journal of the Transportation Research Board, No. 2284, 109–116. Gao, L., Liu, X., Liu, Y., Wang, P., Deng, M., Zhu, Q., and Li, H. (2019). Measuring Road Network Topology Vul- nerability by Ricci Curvature. Physica A: Statistical Mechanics and its Applications, 527. Gill, J. C., and Malamud, B. D. (2014). Reviewing and Visualizing the Interactions of Natural Hazards. Reviews of Geophysics, 52, 680–722. Girgin, S., Necci, A., and Krausmann, E. (2019). Dealing with Cascading Multi-Hazard Risks in National Risk Assessment: The Case of Natech Accidents. International Journal of Disaster Risk Reduction, 35. Hammond, P. (n.d.). Deploying Transportation Resilience Practices in State DOTs. Retrieved from https://apps.trb. org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=4208. Hampton Roads Transportation Planning District Commission. (2014, February). Assessing Vulnerability and Risk of Climate Change Effects on Transportation Infrastructure: Hampton Roads Virginia Pilot. Hillsborough County MPO. (2016). Imagine Hillsborough 2040: Long Range Transportation Plan. Tampa, FL: Hillsborough County Metropolitan Planning Organization. Retrieved from http://www.planhillsborough. org/wp-content/uploads/2014/10/2040-LRTP-Final-Full-Report-revised-3-28-16-1.pdf. Hodgson, D., McDonald, J. L., and Hosken, D. J. (2015). What Do You Mean, ‘Resilient’? Trends in Ecology & Evolution, 30(9), 503–506. Holton, G. A. (2014). Value-at-Risk: Theory and Practice, 2nd ed. Belmont, MA: Glyn A. Holton. Husdal, J. (n.d.). The Vulnerability of Road Networks in a Cost-Benefit Perspective. Retrieved from https://www. researchgate.net/publication/200113562_The_vulnerability_of_road_networks_in_a_cost-benefit_perspective. ICF International. (2014). Assessing Criticality in Transportation Adaptation Planning. Washington, DC: U.S. Department of Transportation. INFRARISK. (n.d.). INFRARISK. Retrieved January 31, 2021, from http://www.infrarisk-fp7.eu/the-project- methodology. Institute for Trade and Transportation Studies. (2020). Freight in the Southeast 10th Conference Trends & Connec- tions: Corridor Planning and Freight Resilience in the Southeast—AASHTO Resilience Connections. Retrieved from https://web.cvent.com/event/203c14a0-fb45-4518-8eea-da55a344a43d/websitePage:645d57e4-75eb- 4769-b2c0-f201a0bfc6ce. International Organization for Standardization (ISO). (2018). ISO 31000 Risk Management: Guidelines. Geneva. IPCC. (2007). AR4 Climate Change 2007: Synthesis Report. (R. K. Pachauri, and A. Reisinger, Eds.) Geneva, Switzerland: Intergovernmental Panel on Climate Change. ISSMGE TC32: Technical Committee on Risk Assessment and Management Glossary of Risk Assessment Terms. (2004). Risk Assessment—Glossary of Terms. Retrieved February 8, 2021, from http://140.112.12.21/ issmge/2004Glossary_Draft1.pdf. Kappes, M. S., Keiler, M., and Glade, T. (2010). From Single- to Multi-Hazard Risk Analysis: A Concept Addressing Emerging Challenges. Proceedings of the Mountain Risks International Conference. Firenze, Italy. KDOT. (2019). Transportation Asset Management Plan. Kansas Department of Transportation.

Literature Review B-71 Kreibich, H., and Bubeck, P. (2013). Natural Hazards: Direct Costs and Losses Due to the Disruption of Production Processes. New York, NY: UNISDR. LaFree, G., Smith, B., Freilich, J., Miller, E., Braniff, W., and Smarick, K. (2019, September). The Terrorism and Extremist Violence in the United States (TEVUS) Database. (START. National Consortium for the Study of Terrorism and Responses to Terrorism.) Retrieved from https://start.umd.edu/research-projects/terrorism- and-extremist-violence-united-states-tevus-database. Litman, T. (2007, January). Evaluating Rail Transit Benefits: A Comment. Transport Policy, 14, 94–97. doi:10.1016/ j.tranpol.2006.09.003. Lu, D., Tighe, S., and Xie, W. (n.d.). Impacts of Flooding on Asphalt Pavements under Climate Change. Retrieved from https://trid.trb.org/View/1511407. Malet, J.-P., and Remaitre, A. (2015). Probabilistic Forecasts of Debris-Flow Hazard at the Regional Scale with a Combination of Models. EGU General Assembly, 17. Markolf, S., Hoehne, C., Fraser, A., Chester, M., and Underwood, B. S. (n.d.). Transportation Resilience to Climate Change and Extreme Weather Events—Beyond Risk and Robustness. Retrieved from https://www.sciencedirect. com/science/article/pii/S0967070X17305000. Massachusetts DOT. (n.d.). Climate Resilience Pilot Project—MassDOT/FHWA. Retrieved January 31, 2021, from https://www.mass.gov/doc/massdot-climate-resilience-modeling-overview-faq/download. Mattsson, L.-G., and Jenelius, E. (2015). Vulnerability and Resilience of Transport Systems—A Discussion of Recent Research. Retrieved from https://www.sciencedirect.com/science/article/abs/pii/S0965856415001603. Merrimack Valley MPO. (2015, July). Merrimack Valley 2016 Regional Transportation Plan. Retrieved from http://www.miamidadetpo.org/library/plans/2040-long-range-transportation-plan-final-2014-10.pdf. Meyer, V., Becker, N., Markantonis, V., and Schwarze, R. (2012). Costs of Natural Hazards—A Synthesis. CONHAZ Consortium. Miami-Dade MPO. (2014, October). Miami-Dade 2040 Long Range Transportation Plan. Retrieved from http:// www.miamidadetpo.org/library/plans/2040-long-range-transportation-plan-final-2014-10.pdf. Mid-Region COG. (2015). Long Range Transportation Plan: Futures 2040 MTP. Albuquerque, NM: Mid-Region Council of Governments. Miller, R. (2018). TAMP: Transportation Asset Management Plan. Washington State Department of Transportation. MnDOT. (2022). Minnesota Go: Planning Minnesota’s Transportation Future. Retrieved from https://www. minnesotago.org/trends. MnDOT. (2018, December). 2017 Minnesota Transportation Results Scorecard. Retrieved January 29, 2021, from https://www.dot.state.mn.us/measures. MnDOT. (2019). Transportation Asset Management Plan. Minnesota Department of Transportation. Moran, M. J., and Monje, C. (2016). Guidance on Treatment of the Economic Value of a Statistical Life (VSL) in U.S. Department of Transportation Analysis—2016 Adjustment. Washington, DC: U.S. Department of Transportation. Retrieved from https://www.transportation.gov/office-policy/transportation-policy/revised-departmental- guidance-on-valuation-of-a-statistical-life-in-economic-analysis. Multi-hazard Loss Estimation Methodology. (n.d.). Retrieved from https://www.fema.gov/sites/default/files/2020-09/ fema_hazus_flood- model_technical-manual_2.1.pdf. National Institute of Standards and Technology (NIST). (n.d.). National Vulnerability Database. Retrieved from https://nvd.nist.gov/. National Research Council. (2010). Adapting to the Impacts of Climate Change, America’s Climate Choices: Panel on Adapting to the Impacts of Climate Change. Washington, DC. https://nap.nationalacademies.org/ read/12783/chapter/1. NOAA. (2022, March 15). NOAA’s National Weather Service Storm Prediction Center. Retrieved from https:// www.spc.noaa.gov/gis/svrgis/. Northern Middlesex MPO. (n.d.). Northern Middlesex Regional Transportation Plan 2016–2040. Lowell, MA: Northern Middlesex Metropolitan Planning Organization. O’Har, J. P., Senesi, C., and Molenaar, K. R. (2016, September). Development of a Risk Register Spreadsheet Tool. NCHRP 08-36/Task 126, Retrieved from https://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp? ProjectID=3685. Oppikoger, T., Hermanns, R. L., Sandoy, G., and Bohme, M. (2016). Quantification of Casualties from Potential Rock-Slope Failures in Norway. In Landslides and Engineered Slopes, Experience, Theory and Practice. Rome, Italy: Associazione Geotechnica Italiana. Palm Beach Metropolitan Planning Organization. (2014). Palm Beach MPO 2040 Long Range Transportation Plan. Palm Beach Metropolitan Planning Organization. Retrieved from https://www.palmbeachtpa.org/ static/sitefiles/LRTP/2040/LRTP.pdf. Parkany, E., and Ogunye, O. (2016). Transportation Corridor Resilience Assessment. Charlottesville, VA: University of Virginia, Mid-Atlantic Transportation Sustainability University Transportation Center.

B-72 Developing a Highway Framework to Conduct an All-Hazards Risk and Resilience Analysis Pearson, D., Stein, S., and Jones, J. S. (2000). Risk-Based Design of Bridge Scour Countermeasures. Transportation Research Record: Journal of the Transportation Research Board, No. 1696, 229–235. PennDOT. (2017). Phase 1 PennDOT Extreme Weather Vulnerability Study Report. Perona, P., Mischer, A., and Porporato, A. (2009). A Probabilistic Model for Snow Avalanche Occurrence. EGU General Assembly Conference Abstracts. Pescaroli, G., and Alexander, D. (2018). Understanding Compound, Interconnected, Interacting and Cascading Risks: A Holistic Framework. Risk Analysis, 38(11), 2245–2257. Pregnolato, M., Ford, A., Wilkinson, S. M., and Dawson, R. J. (n.d.). The Impact of Flooding on Road Trans- port: A Depth-Disruption Function. Retrieved from https://www.sciencedirect.com/science/article/pii/ S1361920916308367. Princeton University. (2016). Chicago Project on Security and Terrorism (CPOST). Suicide Attack Database. Retrieved from https://cpost.uchicago.edu/about/contact/. Proag, V. (2014). The Concept of Vulnerability and Resilience. Procedia Economics and Finance, 18, 369–376. Proctor, G., Varma, S., and Roorda, J. (2016). Managing Risk Across the Enterprise: A Guide for State Depart- ments of Transportation. NCHRP Project 08-93, Final Report. Retrieved from https://onlinepubs.trb.org/ onlinepubs/nchrp/docs/NCHRP08-93_FullGuide.pdf. RAND Corporation. (n.d.). Adding Resilience to the Freight System in Statewide and Metropolitan Transporta- tion Plans: Developing a Conceptual Approach. From NCHRP Project 08-36/Task 73. Retrieved from http:// onlinepubs.trb.org/onlinepubs/nchrp/docs/NCHRP08-36(73)_FR.pdf. RAND Corporation. (2018, September 11). RAND Database of Worldwide Terrorism Incidents. Retrieved from http://smapp.rand.org/rwtid/search_form.php. Regional Planning Commission. (2015, January). Metropolitan Transportation Plan: New Orleans Urbanized Area. Retrieved from http://www.norpc.org/assets/pdf-documents/2044%20NO%20MTP%20FINAL%20 ADOPTED.pdf. Resilience. (n.d.). Retrieved from https://www.fhwa.dot.gov/environment/sustainability/resilience/. Risk Management Guidance. (n.d.). Retrieved from https://highways.dot.gov/sites/fhwa.dot.gov/files/docs/ federal-lands/construction/15931/risk-management-guidance.pdf. Sanford, W. (2015). An Overview of ISO 55000—Standardizing Asset Management. Retrieved from http:// pinnaclereliability.com/wp-content/uploads/2021/01/IJ_Nov-Dec-2015_An-Overview-of-ISO55000.pdf. Serulle, N. U., Heaslip, K., and Brady, B. (2011). A Case Study of Network Transportation Resilience of Santo Domingo, Dominican Republic. Transportation Research Record: Journal of the Transportation Research Board, No. 2234, 22–30. Smith, B. L., and Damphousse, K. R. (2007, July 30). American Terrorism Study, 1980–2002 (ICPSR 4639). Retrieved from https://www.icpsr.umich.edu/web/NACJD/studies/4639. Souppaya, M., and Scarfone, K. (2016). Guide to Data-Centric System Threat Modeling. Gaithersburg, MD: National Institute for Standards and Technology. Southeast Michigan Council of Governments (SEMCOG) and Michigan DOT (MDOT). (2020). Climate Resil- iency and Flooding Mitigation Study. Retrieved from https://semcog.org/Portals/0/Documents/Plans-For- The-Region/Environment/SEMCOG%20Climate%20Resiliency%20and%20Flooding%20Mitigation%20 Study_Report_August%202020.pdf?ver=pjn6fTnLv9BZaM8MuasqVw%3d%3d. Stanujkic, D., and Zavadskas, E. (2015). A Modified Weighted Sum Method Based on the Decision-maker’s Preferred Levels of Performances. Studies in Informatics and Control, 24(4). Stein, S., and Sedmera, K. (2006). NCHRP Web-Only Document 107: Risk-Based Management Guidelines for Scour at Bridges with Unknown Foundations. TRB, National Research Council, Washington, DC. Sullivan, J. L., Novak, D. C., Aultman-Hall, L., and Scott, D. M. (n.d.). Identifying Critical Road Segments and Measuring System-Wide Robustness in Transportation Networks with Isolating Links: A Link-Based Capacity-Reduction Approach. Retrieved from https://www.sciencedirect.com/science/article/abs/pii/ S0965856410000418. Sun, W., Bocchini, P., and Davison, B. D. (2018). Resilience Metrics and Measurement Methods for Transporta- tion Infrastructure: The State of the Art. Sustainable and Resilient Infrastructure, 5, 168–199. Ta, C., Goodchild, A. V., and Pitera, K. (2009). Structuring a Definition of Resilience for the Freight Transporta- tion System. Transportation Research Record: Journal of the Transportation Research Board, No. 2097, 19–25. doi:10.3141/2097-03. The National Academies. (2012). Disaster Resilience: A National Imperative. National Academies Press. Retrieved from https://nap.nationalacademies.org/catalog/13457/disaster-resilience-a-national-imperative. Tri-County Regional Planning Commission. (2015). Envision HOI: Heart of Illinois Long Range Transportation Plan. Peoria, IL: Tri-County Regional Planning Commission. Turner, D. (2015). Fragility Assessment of Bridge Superstructures under Hydrodynamic Forces. Retrieved from https:// mountainscholar.org/bitstream/handle/10217/170400/Turner_colostate_0053N_13377.pdf?sequence=1. UDOT. (2019). Utah Transportation Asset Management Plan. Salt Lake City, UT: State of Utah.

Literature Review B-73 UNISDR. (2004). Living with risk: a global review of disaster reduction initiatives. Geneva: UN Publications. University of Maryland. (2020). Global Terrorism Database. Retrieved from https://start.umd.edu/gtd/. U.S. DOT. (n.d.). FY-2015 Annual Performance Report/FY-2017 Annual Performance Plan. Washington, DC. Retrieved from https://www7.transportation.gov/mission/budget/fy2017-annual-performance-plan-fy- 2015-perf-report. USACE. (2017a). Sea-Level Change Curve Calculator (Version 2019.21). Retrieved from https://cwbi-app.sec. usace.army.mil/rccslc/slcc_calc.html. USACE. (2017b). Sea-Level Tracker. Retrieved from https://climate.sec.usace.army.mil/slr_app/. USACE. (n.d.a). HEC-RAS. Retrieved from https://www.hec.usace.army.mil/software/hec-ras/. USACE. (n.d.b). Public Tools Developed by USACE—Climate Impacted Hydrology. Retrieved from https://www. usace.army.mil/corpsclimate/Public_Tools_Dev_by_USACE/Climate-Impacted_Hydrology.aspx. USGAO. (2020, August 5). A Climate Migration Pilot Program Could Enhance the Nation’s Resilience and Reduce Federal Fiscal Exposure. Retrieved from https://www.gao.gov/products/gao-20-488. USGS. (1971, February 9). CISN Rapid Instrumental Intensity Map for San Fernando Earthquake. Retrieved from https://www.cisn.org/shakemap/sc/shake/San_Fernando/intensity.html. USGS. (n.d.a). Landslide Inventory. Retrieved from https://usgs.maps.arcgis.com/apps/webappviewer/index.htm l?id=ae120962f459434b8c904b456c82669d. USGS. (n.d.b). Peak Streamflow for the Nation. Retrieved August 12, 2020, from https://nwis.waterdata.usgs.gov/ usa/nwis/peak. van Westen, C., Kingma, N., Damen, M., Kerle, N., and Alkema, D. (2011). Multi-Hazard Risk Assessment: Distance Education Course Guide Book. Twente, Netherlands: University of Twente. VizSec Organization. (n.d.). VIZSEC. Retrieved from https://vizsec.org/data/ Volpe. (2013, September 17). Infrastructure Resiliency: A Risk-Based Framework. Retrieved January 15, 2020, from https://www.volpe.dot.gov/sites/volpe.dot.gov/files/docs/Infrastructure%20Resiliency_A%20Risk- Based%20Framework.pdf. VTRANS. (n.d.). Transportation Resilience {Planning Tool). Retrieved January 31, 2021, from https://vtrans. vermont.gov/planning/transportation-resilience. VTRANS. (2019.) Why Should We Care About Roads? Retrieved from https://vtrans.vermont.gov/sites/aot/files/ planning/documents/TAMP%20Brochure-SCROLLING.pdf. Weilant, S., Strong, A., and Miller, B. (2019). Incorporating Resilience into Transportation Planning and Assess- ment. Santa Monica, CA: RAND Corporation. Retrieved from https://www.rand.org/content/dam/rand/ pubs/research_reports/RR3000/RR3038/RAND_RR3038.pdf. Werner, S., et al. (2006). REDARS 2 Methodology and Software for Seismic Risk Analysis of Highway Systems. U.S. Department of Transportation, Federal Highway Administration. WHO. (2009). Qualitative Risk Characterization in Risk Assessment. In Risk Characterization of Microbiological Hazards in Food: Guidelines, Vol. 17. Winter, M. G., Smith, J. T., Fotopoulou, S., Pitilakis, K., Mavrouli, O., Corominas, J., and Argyroudis, S. (2014). An Expert Judgement Approach to Determining the Physical Vulnerability of Roads to Debris Flow. Bulletin of Engineering Geology and the Environment, 73(2), 291–305. WSDOT. (2011, November). Climate Impacts Vulnerability Assessment. Retrieved from https://wsdot.wa.gov/ sites/default/files/2017/11/15/ENV-Climate-VulnerabilityAssessment.pdf. WSP. (n.d.). Resiliency Case Studies: State DOT Lessons Learned. Retrieved from https://environment.transportation. org/pdf/rsts/aashto_resiliency%20_case_studies.pdf WV DOT Transportation Asset Management Plan. (2019). Retrieved from https://transportation.wv.gov/highways/ programplanning/Documents/2019-Final-TAMP.pdf. Wylie, D. C. (2015). Rock Fall Engineering. Boca Raton, FL: CRC Press. Zapata, C., and Houston, W. (2008). NCHRP Report 602: Calibration and Validation of the Enhanced Integrated Climatic Model for Pavement Design. Transportation Research Board of the National Academies, Washington, DC. Retrieved from https://www.nap.edu/catalog/23098/calibration-and-validation-of-the-enhanced- integrated-climatic-model-for-pavement-design.